CN108633249A

CN108633249A - A kind of physiological signal Quality estimation method and device

Info

Publication number: CN108633249A
Application number: CN201780009418.0A
Authority: CN
Inventors: 张�杰; 陈文娟; 董辰; 朱萸
Original assignee: Huawei Technologies Co Ltd
Current assignee: Huawei Technologies Co Ltd
Priority date: 2017-01-25
Filing date: 2017-05-27
Publication date: 2018-10-09
Anticipated expiration: 2037-05-27
Also published as: CN108633249B; WO2018137300A1

Abstract

A kind of physiological signal Quality estimation method and device, is related to wearable device field, can improve the accuracy of physiological signal Quality estimation result, and reduces the calculation amount during physiological signal Quality estimation.Concrete scheme is：S301, acquisition physiological signal, the physiological signal are periodic signal or class periodic signal；S302, the characteristic point for extracting physiological signal, this feature point includes the characteristic point for being used to indicate the physiological signal period；S303, the characteristic point according to physiological signal divide the period to the physiological signal；S304, judged according to the similitude between the signal and current demand signal template set in i-th of period i-th of period signal signal quality；Wherein, signal templates set includes N number of signal templates, and N number of signal templates are obtained according to the physiological signal before the signal in i-th of period, and N is more than or equal to 2, i and is more than N.This method and device according to the physiological signal of human body for before obtaining Human Physiology information, during judging the physiological signal quality.

Description

Physiological signal quality judgment method and device

The present application claims priority of chinese patent application entitled "method and system for real-time determination of physiological signal quality with extremely low resource consumption" filed by chinese patent office on 25/01/2017 with application number 201710061354.0, the entire contents of which are incorporated herein by reference.

Technical Field

The embodiment of the invention relates to the field of wearable equipment, in particular to a method and a device for judging the quality of a physiological signal.

Background

In recent years, wearable devices are increasingly applied to physiological monitoring of the human body due to their characteristics of being non-invasive, simple and flexible. The wearable device can provide low-load, non-contact and long-term continuous physiological monitoring for a patient, such as monitoring physiological signals of Electrocardio (ECG), pulse wave (PPG), respiration, blood pressure and the like of a human body. Wherein, the pulse wave is short for the plethysmography information of the electric volume. The current wearable equipment acquires the physiological signals of a human body through the wearable sensor, and the wearable sensor is easily interfered by noise and motion artifacts, so that the acquired physiological signals and the physiological information of the human body obtained based on the acquired physiological signals deviate from the real situation. Thus, before obtaining physiological information of a human body using a physiological signal, the signal quality of the physiological signal must be judged. Specifically, it is determined whether the physiological signal is a normal physiological signal or an abnormal physiological signal interfered by noise and motion artifacts.

In the prior art, before the quality of the obtained physiological signals of the electrocardio, the pulse wave, the respiration, the blood pressure and the like of the human body is judged, a signal template for judging the quality of the physiological signals is generally required to be preset; in addition, in the process of acquiring the physiological signal, a large number of characteristic values of the physiological signal generally need to be accurately acquired, that is, a large number of characteristic points of the physiological signal are accurately extracted; therefore, the quality of the physiological signal is judged according to the preset signal template and the large number of accurate characteristic values.

The method for determining the quality of the physiological signals of the patient according to the preset signal template has the problems that the preset signal template is usually obtained from a large number of off-line physiological signals by adopting a machine learning algorithm or priori knowledge, and the off-line physiological signals have certain deviation from the physiological signals obtained in real time, so that the accuracy of the quality determination result of the physiological signals obtained according to the preset signal template needs to be improved. Moreover, because the current wearable sensor is generally simpler, some features of the physiological signal, such as the dicrotic wave crest, may not be extracted, i.e., the large number of accurate feature values may not be obtained necessarily; therefore, the accuracy of the physiological signal quality judgment result obtained according to the large number of accurate characteristic values needs to be improved, and the calculation amount in the process of judging the physiological signal quality is large.

Disclosure of Invention

The application provides a method and a device for judging the quality of a physiological signal, which can improve the accuracy of a judgment result of the quality of the physiological signal and reduce the calculation amount in the judgment process of the quality of the physiological signal.

In order to achieve the purpose, the technical scheme is as follows:

in a first aspect, a method for determining a quality of a physiological signal is provided, the method comprising: collecting a physiological signal, wherein the physiological signal is a periodic signal or a quasi-periodic signal; extracting feature points of a physiological signal, wherein the feature points comprise feature points used for indicating the period of the physiological signal; dividing a period of a physiological signal according to the characteristic points of the physiological signal; judging the signal quality of the signal of the ith period according to the similarity between the signal of the ith period and the current signal template set; the signal template set comprises N signal templates, the N signal templates are obtained according to physiological signals before the signal of the ith period, N is greater than or equal to 2, and i is greater than N. Generally, each signal template of the N signal templates in the current signal template set is a good-quality physiological signal. Therefore, if the similarity between the signal of the ith period and the current signal template set is higher, the signal quality of the signal of the ith period is better; the lower the similarity between the signal of the ith period and the current set of signal templates, the worse the signal quality of the signal of the ith period.

It should be noted that, in the physiological signal quality determination method provided by the present application, the current signal template set for determining the physiological signal quality is obtained from the physiological signals acquired in real time, rather than from a large number of off-line physiological signals by using a machine learning algorithm or priori knowledge; therefore, the current signal template provided by the application is more consistent with the physiological signal acquired in real time, and the accuracy of the physiological signal quality judgment result obtained according to the current signal template set is higher.

In a possible implementation manner, the signal quality of the signal of the ith period can be determined by determining whether the signal of the ith period meets the standard of a normal physiological signal, that is, whether the signal of the ith period is a normal physiological signal or an abnormal physiological signal. Specifically, the above determining the signal quality of the signal in the ith period according to the similarity between the signal in the ith period and the signal template set may include: determining a similarity result of the signal of the ith period, wherein the similarity result of the signal of the ith period is used for indicating the similarity between the signal of the ith period and the current signal template set; if the similarity result of the signal of the ith period meets a preset similarity condition, judging the signal of the ith period to be a normal physiological signal, wherein the preset similarity condition is preset; and if the similarity result of the signal of the ith period does not meet the preset similarity condition, judging the signal of the ith period as an abnormal physiological signal.

The similarity between the signal of the ith period and the current signal template set may be represented by similar parameters such as a correlation coefficient and a mean square error between the sample sequence of the signal of the ith period and the sample sequence of the signal template. For example, in the case where the similarity parameter of the signal of the i-th period is the correlation coefficient, the larger the similarity parameter of the signal of the i-th period is, the higher the similarity is, and the better the signal quality is. The smaller the similarity parameter of the signal of the ith period is, the lower the similarity is, and the worse the signal quality is.

In a possible implementation manner, in a case that it is determined that the similarity result of the signal of the ith period is better than the similarity result of at least one signal template of the N signal templates, the method may further include: and updating the current signal template set according to the signal of the ith period. That is, if the signal quality of the signal of the ith period is higher than the signal quality of at least one of the N signal templates in the current signal template set, the method may replace the signal template according to the signal of the ith period to update the current signal template set. Wherein, the similarity result of one signal template in the N signal templates is obtained according to the physiological signal before the signal of the ith period.

It should be noted that the current signal template set is continuously updated according to the physiological signals acquired in real time. And replacing the signal template with relatively poor signal quality according to the periodic signal with better signal quality to update the current signal template set, so that the signal quality of the signal template in the current signal template set is improved in real time. Therefore, the accuracy of the physiological signal quality judgment result obtained according to the current signal template set updated in real time is higher.

In a possible implementation manner, since the signal quality of the signal of the ith period is related to the similarity result of the signal of the ith period, the current signal template set may be updated according to the similarity result of the signal of the ith period. Specifically, the updating the current signal template set according to the signal of the ith period may include: in the event that it is determined that the similarity result of the signal of the i-th cycle is better than the similarity result of at least one of the N signal templates: and replacing the signal template with the signal with the worst similar result in the current signal template set by using the signal of the ith period.

Alternatively, the above-mentioned "replacing the signal template with the signal of the ith period for the signal template with the worst similar result in the current signal template set" may be replaced by replacing any signal template with the signal of the ith period for the signal template with the worst similar result in the current signal template set. Wherein, the similarity result of any signal template with poor similarity result is worse than that of the signal of the ith period.

In a possible implementation manner, before determining the signal quality of the signal of the ith period according to the similarity between the signal of the ith period and the current signal template set, the method may further include: and acquiring a current signal template set according to the physiological signal before the signal of the ith period.

The current signal template set may be a signal template set that has been updated, for example, the current signal template set may be a signal template set that is updated according to the signal of the i-1 th cycle. Alternatively, the current set of signal templates may be a set of signal templates that have not been updated.

In a possible implementation manner, the obtaining the current signal template set according to the physiological signal before the ith cycle of signal may include: acquiring an initial signal template set according to the signals of the first N + a periods; and acquiring a current signal template set according to the initial signal template set and physiological signals from the signal of the N + a th period to the signal of the ith period. The initial signal template set is a signal template set which is not updated.

It should be noted that the N signal templates in the initial signal template set are not necessarily all normal physiological signals with good signal quality, but the physiological signal quality determination method may be executed as long as the initial signal template set includes the N signal templates. The method can judge the signal quality of each period of signal according to the similarity result of each period of signal from the signal of the (N + a) th period to the signal of the (i) th period while acquiring the current signal template set according to the physiological signal between the initial signal template set and the signal of the (N + a) th period to the signal of the (i) th period.

In a possible implementation manner, the acquiring an initial signal template set according to the first N + a cycles of signals may include: determining a similarity result of the signal of each period in the signals of the first N + a periods; a is an integer greater than or equal to 0, and N + a is less than i; wherein, the similarity result of the signal of one period in the first N + a periodic signals is composed of the similarity between the signal of the period and the signals of other periods except the signal of the period in the first N + a periodic signals; and determining the signals of the first N periods with the best similar result in the signals of the first N + a periods as an initial signal template set.

When a is equal to 0, the N signal templates in the initial signal template set are the first N periods of signals. When a is larger than 0, the signals of the first N + a periods have a periods which are not the signal templates in the initial signal template set. It should be noted that, after the similarity of the signal in each period of the first N + a periods is determined, the signal quality of the signal in each period of the first N + a periods may also be determined.

In a possible implementation manner, the extracting the feature points of the physiological signal may further include: acquiring a characteristic value of the signal of the ith period according to the characteristic point of the physiological signal; determining that the characteristic value of the signal of the ith period is not in the current preset threshold range, and judging that the signal of the ith period is an abnormal physiological signal, wherein the current preset threshold range is determined according to the signal of the period before the signal of the ith period; before determining the signal quality of the signal of the ith period according to the similarity between the signal of the ith period and the signal template set, the method may further include: and determining that the characteristic value of the signal of the ith period is in the current preset threshold range.

It should be noted that, the current preset threshold range is obtained according to signals of all cycles in the real-time acquired physiological signals, that is, the current preset threshold range is obtained according to a global variable; therefore, the current preset threshold range conforms to the physiological signal acquired in real time. Therefore, the physiological signal quality judgment method provided by the application can further improve the accuracy of the physiological signal quality judgment result.

In a possible implementation manner, in a case where the extracted feature points of the physiological signal include a top point and a bottom point, the feature value of the signal of the i-th cycle includes: a period value of the signal of the ith period and/or a height value of the signal of the ith period. Thus, the current preset threshold range may include: a current preset period range and/or a current preset height range. Specifically, the step of determining that the characteristic value of the signal in the ith period is not within the current preset threshold range may include: the period value of the signal of the ith period is not in the current preset period range, and/or the height value of the signal of the ith period is not in the current preset height range. The characteristic value of the signal of the ith period in the current preset threshold range may include: the period value of the signal of the ith period is in a current preset period range and the height value of the signal of the ith period is in a current preset height range.

It should be noted that the method provided by the present application can determine the signal quality of the signal in each period of the physiological signal by extracting the top point and the bottom point in the characteristic value of the physiological signal, and can reduce the amount of calculation in the process of determining the physiological signal quality to a certain extent.

In a possible implementation manner, before the obtaining a feature value of the signal of the i-th cycle according to the feature point of the physiological signal, the method may further include: determining a current preset threshold range according to a signal of a period before the signal of the ith period; after obtaining the characteristic value of the signal of the ith cycle according to the characteristic point of the physiological signal, the method further comprises: and updating the current preset threshold range according to the characteristic value of the signal of the ith period.

The current preset threshold range is continuously updated according to the physiological signals acquired in real time; the current preset threshold range is therefore consistent with the physiological signals acquired in real time. Therefore, the physiological signal quality judgment method provided by the application can further improve the accuracy of the physiological signal quality judgment result.

In a possible implementation manner, the method may further include: and outputting a signal quality judgment result, wherein the signal quality judgment result comprises that the signal of the ith period is a normal physiological signal or the signal of the ith period is an abnormal physiological signal, and the signal of each period before the signal of the ith period is a normal physiological signal or the signal of each period is an abnormal physiological signal. If the similarity result of the signal in a period before the signal in the ith period meets the preset similarity condition, the signal in the period is a normal physiological signal; and if the similarity result of the signal in one period before the signal in the ith period does not meet the preset similarity condition, the signal in the period is an abnormal physiological signal.

It should be noted that the method for judging the quality of the physiological signal provided by the present application can judge the signal quality of the acquired physiological signal cycle by cycle in real time, and can distinguish whether the signal of any cycle in the physiological signal is a normal physiological signal or an abnormal physiological signal more accurately. Therefore, more accurate human physiological information can be obtained according to the normal physiological signals obtained by judgment. And the signal quality judgment result is output, so that the result can be visually displayed to a user or related technical personnel, and the user experience is improved or the requirements of the related technical personnel are met.

In a second aspect, a physiological signal quality determination apparatus is provided, including: the device comprises an acquisition module, an extraction module, a division module and a judgment module. The acquisition module is used for acquiring physiological signals, and the physiological signals are periodic signals or quasi-periodic signals. And the extraction module is used for extracting the characteristic points of the physiological signals acquired by the acquisition module, wherein the characteristic points comprise characteristic points used for indicating the period of the physiological signals. And the dividing module is used for dividing the physiological signal into cycles according to the characteristic points of the physiological signal extracted by the extracting module. The judging module is used for judging the signal quality of the signal of the ith period according to the similarity between the signal of the ith period and the current signal template set; the signal template set comprises N signal templates, the N signal templates are obtained according to physiological signals before the ith period of signals acquired by the acquisition module, N is greater than or equal to 2, and i is greater than N.

In a possible implementation manner, the apparatus may further include: and determining a module. The determining module is used for determining a similarity result of the signal of the ith period divided by the dividing module, and the similarity result of the signal of the ith period is used for indicating the similarity between the signal of the ith period and the current signal template set. The determining module is specifically configured to determine that the signal of the ith cycle is a normal physiological signal if the similarity result of the signal of the ith cycle meets a preset similarity condition, where the preset similarity condition is preset; and if the similarity result of the signal of the ith period does not meet the preset similarity condition, judging the signal of the ith period as an abnormal physiological signal.

In a possible implementation manner, the apparatus may further include: and updating the module. The updating module is used for updating the current signal template set according to the signal of the ith period under the condition that the similarity result of the signal of the ith period is better than the similarity result of at least one signal template in the N signal templates, wherein the similarity of one signal template in the N signal templates is obtained according to the physiological signal before the signal of the ith period.

In a possible implementation manner, the updating module is specifically configured to replace the signal template with the worst similar result in the current signal template set by the signal of the ith period.

In a possible implementation manner, the apparatus may further include: and an acquisition module. The acquisition module is used for acquiring the current signal template set according to the physiological signal before the signal of the ith period before the judgment module judges the signal quality of the signal of the ith period according to the similarity between the signal of the ith period and the current signal template set.

In a possible implementation manner, the obtaining module is specifically configured to obtain an initial signal template set according to the signals of the first N + a cycles; and acquiring a current signal template set according to the initial signal template set and physiological signals from the signal of the N + a th period to the signal of the ith period.

In a possible implementation manner, the determining module may be further configured to determine similarity of signals in each of the first N + a periods of the signals; a is an integer greater than or equal to 0, and N + a is less than i; wherein, the similarity result of the signal of one period in the first N + a period signals is composed of the similarity between the signal of the period and the signals of other periods except the signal of the period in the first N + a period signals. The obtaining module is specifically configured to determine, as an initial signal template set, a signal of a first N cycles with a best similarity result among the signals of the first N + a cycles determined by the determining module.

In a possible implementation manner, the obtaining module may be further configured to, after the extracting module extracts the feature points of the physiological signal, obtain the feature value of the signal of the ith cycle according to the feature points of the physiological signal. The judging module may be further configured to determine that the characteristic value of the signal in the ith period is not within a current preset threshold range, and judge that the signal in the ith period is an abnormal physiological signal, where the current preset threshold range is determined according to the signal in the period before the signal in the ith period; before the signal quality of the signal of the ith period is judged according to the similarity between the signal of the ith period and the current signal template set, the characteristic value of the signal of the ith period is determined to be in the current preset threshold range.

In a possible implementation manner, the characteristic value of the signal of the i-th cycle includes: a period value of the signal of the ith period and/or a height value of the signal of the ith period. The current preset threshold range may include: a current preset period range and/or a current preset height range. The characteristic value of the signal of the ith period not being within the current preset threshold range may include: the period value of the signal of the ith period is not in the current preset period range, and/or the height value of the signal of the ith period is not in the current preset height range. The characteristic value of the signal of the ith period in the current preset threshold range may include: the period value of the signal of the ith period is in a current preset period range, and the height value of the signal of the ith period is in a current preset height range.

In a possible implementation manner, the determining module may be further configured to determine the current preset threshold range according to the signal of the cycle before the signal of the ith cycle before the obtaining module obtains the feature value of the signal of the ith cycle according to the feature point of the physiological signal. The updating module may be further configured to update the current preset threshold range according to the characteristic value of the signal in the ith period after the obtaining module obtains the characteristic value of the signal in the ith period according to the characteristic point of the physiological signal.

In a possible implementation manner, the apparatus may further include: and an output module. The output module is used for outputting a signal quality judgment result, wherein the signal quality judgment result comprises that the signal of the ith period obtained by the judgment module is a normal physiological signal or the signal of the ith period is an abnormal physiological signal, and the signal of each period before the signal of the ith period is a normal physiological signal or the signal of each period is an abnormal physiological signal. If the similarity result of the signal in a period before the signal in the ith period meets a preset similarity condition, judging the signal in the period to be a normal physiological signal; and if the similarity result of the signal in one period before the signal in the ith period does not meet the preset similarity condition, judging the signal in the period to be an abnormal physiological signal.

In a third aspect, a physiological signal quality determination apparatus may include a processor, a memory, a display, an input, and a bus; the memory is used for storing the at least one instruction, the processor, the memory, the display and the input device are connected through the bus, and when the apparatus is operated, the processor executes the at least one instruction stored in the memory, so that the apparatus executes the physiological signal quality judging method in the first aspect and the various optional modes of the first aspect.

In a fourth aspect, a computer storage medium having at least one instruction stored therein is provided; the at least one instruction, when executed on the computer, causes the computer to perform a method of physiological signal quality determination as in the first aspect and various alternatives of the first aspect.

In a fifth aspect, a computer program is provided, having at least one instruction stored therein; the at least one instruction, when executed on the computer, causes the computer to perform a method of physiological signal quality determination as in the first aspect and various alternatives of the first aspect.

It should be noted that, the processor in the third aspect of the present application may be an integration of the functional modules in the second aspect, such as the extracting module, the dividing module, the judging module, the determining module, the updating module, and the obtaining module, and the processor may implement the functions of the above functional modules in the second aspect. For the detailed description and the beneficial effect analysis of each module in the second aspect and the third aspect, reference may be made to the corresponding description and the technical effect in the first aspect and various possible implementation manners thereof, which are not described herein again.

Drawings

Fig. 1 is a schematic structural diagram of a wearable device according to an embodiment of the present invention;

FIG. 2 is a schematic waveform diagram of a physiological signal according to an embodiment of the present invention;

fig. 3 is a schematic flowchart of a method for determining a quality of a physiological signal according to an embodiment of the present invention;

FIG. 4 is a schematic waveform diagram of another physiological signal according to an embodiment of the present invention;

FIG. 5 is a schematic waveform diagram of another physiological signal according to an embodiment of the present invention;

fig. 6 is a schematic flowchart of another method for determining physiological signal quality according to an embodiment of the present invention;

fig. 7 is a flowchart illustrating another method for determining physiological signal quality according to an embodiment of the present invention;

FIG. 8 is a flowchart illustrating another method for determining a physiological signal quality according to an embodiment of the present invention;

FIG. 9 is a flowchart illustrating another method for determining physiological signal quality according to an embodiment of the present invention;

FIG. 10 is a flowchart illustrating another method for determining physiological signal quality according to an embodiment of the present invention;

fig. 11 is a flowchart illustrating another method for determining physiological signal quality according to an embodiment of the present invention;

fig. 12 is a flowchart illustrating another method for determining physiological signal quality according to an embodiment of the present invention;

fig. 13 is a flowchart illustrating another method for determining physiological signal quality according to an embodiment of the present invention;

fig. 14 is a flowchart illustrating another method for determining physiological signal quality according to an embodiment of the present invention;

FIG. 15 is a schematic diagram of a possible component of a physiological signal quality determining apparatus according to an embodiment of the present invention;

fig. 16 is a schematic diagram of another possible composition of a physiological signal quality determining apparatus according to an embodiment of the present invention;

fig. 17 is a schematic diagram of another possible composition of a physiological signal quality determining apparatus according to an embodiment of the present invention;

fig. 18 is a schematic diagram of another possible composition of a physiological signal quality determining apparatus according to an embodiment of the present invention;

fig. 19 is a schematic diagram of another possible composition of a physiological signal quality determining apparatus according to an embodiment of the present invention;

fig. 20 is a schematic diagram of another possible composition of a physiological signal quality determining apparatus according to an embodiment of the present invention.

Detailed Description

The embodiment of the invention provides a method and a device for judging the quality of a physiological signal, which are applied to the process of obtaining human physiological information according to the physiological signal of a human body, in particular to the process of judging the quality of the physiological signal before the human physiological information is obtained according to the physiological signal of the human body, so that the accuracy of the judgment result of the quality of the physiological signal can be improved, and the calculation amount in the judgment process of the quality of the physiological signal can be reduced.

The technical solutions in the embodiments of the present invention will be described in detail below with reference to the accompanying drawings in the embodiments of the present invention.

In the method for judging the quality of a physiological signal according to the embodiment of the present invention, the physiological signal may be a periodic or quasi-periodic physiological signal, such as a periodic physiological signal of an electrocardiogram, a pulse wave, a respiration, a blood pressure, and the like of a human body. The physiological signals all contain a large amount of human physiological information, for example, the pulse wave signals of the human body can contain the heart rate, the respiratory rate, the blood oxygen and other physiological information of the human body.

It should be noted that, the method for determining the quality of a physiological signal according to the embodiment of the present invention may determine, in addition to the signal quality of the physiological signal, the signal quality of other periodic or quasi-periodic signals, such as an acceleration signal in a step counting algorithm, which is not limited in the embodiment of the present invention.

The physiological signal quality judgment device provided by the embodiment of the invention can be wearable equipment capable of acquiring physiological signals of human bodies. The wearable device is directly worn on the body of a user or is a portable device integrated into clothes or accessories of the user, most of the wearable devices are internally provided with intelligent systems and can be connected with mobile phones and various terminals, and the wearable devices have one or more functions of photographing, GPS positioning, in-person conversation, intelligent anti-loss, sleep monitoring, heart rate monitoring, running and step recording. Common wearable equipment includes products such as intelligent bracelet, intelligent wrist-watch that use the wrist as the support to wear the product etc. on intelligent shoes, socks or other legs that the foot is the support, products such as intelligent glasses, helmet, the bandeau that use the head as the support, and the products of all kinds of forms such as intelligent body temperature subsides, rhythm of the heart area, intelligent clothing, schoolbag, walking stick, accessory.

For example, fig. 1 is a schematic structural diagram of a wearable device provided in an embodiment of the present invention, and referring to fig. 1, a wearable device 10 may include one or more of the following components: sensor component 101, memory 102, processing component 103, multimedia component 104, audio component 105, interface to input/output (I/O) 106, power component 107, and communication component 108.

The sensor assembly 101 includes one or more sensors for providing various aspects of status assessment for the wearable device 10. The state change of the various aspects provided by the wearable device 10 may be caused by an operation of a user using the wearable device 10 or a change in physiological information of the user.

For example, the sensor assembly 101 may detect the open/closed state of the wearable device 10, the relative positioning of the assemblies. The sensor assembly 101 may also detect a change in the position of the wearable device 10 or one of the components, an orientation or acceleration/deceleration of the wearable device 10, and a change in the temperature of the wearable device 10. The sensor assembly 101 may include a proximity sensor configured to detect the presence of a nearby object without any physical contact. Specifically, the sensor assembly 101 may include a pulse sensor, such as an infrared pulse sensor, a heart rate pulse sensor, a photoelectric pulse sensor, a wrist pulse sensor, or a digital pulse sensor, for detecting a pulse wave of a user, so that physiological information such as a heart rate of a human body can be obtained. A commonly used pulse sensor may be a PPG sensor, among others. The sensor assembly 101 may include a blood pressure sensor for detecting the blood pressure of a human body. In some embodiments, the sensor assembly 101 may also include a light sensor, such as a CMOS or CCD image sensor, for use in imaging applications. For example, the CMOS or CCD image sensor can be used to acquire a video signal of an exposed skin portion such as a face or a hand of a human body. The video signal may include a plurality of frames, each frame is a two-dimensional image, the two-dimensional image is a red, green and blue RGB image, the RGB image may be divided into three images, i.e., an R image, a G image, and a B image, and the three images may be used to extract a pulse wave signal of a human body. In other embodiments, the sensor assembly 101 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

The memory 102 is configured to store various types of data to support operation at the wearable device 10. Examples of such data include instructions for any application or method operating on the wearable device 10, contact data, phone book data, messages, pictures, videos, and human physiological information, among other data. The memory 102 may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.

The processing component 103 generally controls the overall operation of the wearable device 10, such as operations associated with display, phone calls, data communications, camera operations, and recording operations, as well as processing signals or data acquired by sensors. Specifically, after the sensor assembly 101 detects and acquires a physiological signal, the processing assembly 103 may determine the quality of the physiological signal, and obtain that the physiological signal is a normal physiological signal or that the physiological signal is an abnormal physiological signal; thus, human physiological information is obtained according to the normal physiological signals. Among other things, processing component 103 may include one or more processors 1031 to execute instructions. Further, the processing component 103 can include one or more modules that facilitate interaction between the processing component 103 and other components. For example, the processing component 103 may include a multimedia module to facilitate interaction between the multimedia component 104 and the processing component 103.

The multimedia component 104 includes a screen that provides an output interface between the wearable device 10 and the user. In some embodiments, the screen may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of the touch or slide action but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 104 includes a camera. When the wearable device 10 is in an operating mode, such as a shooting mode or a video mode, the camera may receive external multimedia data. Each camera may be a fixed optical lens system or have a focal length and optical zoom capability. Of course, the video signal for acquiring the skin exposed portion of the face, the hand, and the like of the human body may be acquired by a camera in the multimedia component 104, in addition to the sensor such as the CMOS or CCD image sensor, which is not limited in the embodiment of the present invention.

The audio component 105 is configured to output and/or input audio signals. For example, the audio component 105 includes a Microphone (MC) configured to receive external audio signals when the wearable device 10 is in an operating mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signals may further be stored in the memory 102 or transmitted via the communication component 108. In some examples, audio component 105 also includes a speaker for outputting audio signals. Specifically, the audio component 105 may be configured to output the judgment result of the quality of the physiological signal judged by the processing component 103, and output human physiological detail information obtained from the physiological signal, such as a heart rate of a human being of 65 beats per minute (bpm).

The I/O interface 106 provides an interface between the processing component 102 and peripheral interface modules, which may be click wheels, buttons, and the like. These buttons may include, but are not limited to: a home button, a start button, and a lock button. In particular, in the implementation of the present invention, the home button in the I/O interface 106 can also be used to instruct the processing component 103 to start processing the physiological signals acquired by the sensor component 101.

The power component 107 provides power to the various components of the wearable device 10. Power components 107 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for wearable device 10.

The communication component 108 may support wired or wireless communication between the wearable device 10 and other devices such that the wearable device 10 may access a wireless network based on a communication standard, such as WF, 2G, or 3G, or a combination thereof. In an exemplary embodiment, the communication component 108 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 108 further includes a Near Field Communication (NFC) module to facilitate short-range communication. For example, the NFC module may be implemented based on radio frequency identification (RFD) technology, infrared data association (rDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies. Specifically, in this embodiment of the present invention, the communication component 108 may be configured to send the physiological signal quality judgment result obtained by the processing component 103 to other devices, or send human physiological information obtained according to the physiological signal, for example, the heart rate of the human body is 65 bpm.

In an example embodiment, the wearable device may be implemented by one or more application specific integrated circuits (ASCs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components.

In an exemplary embodiment, a non-transitory computer-readable storage medium is also provided that includes instructions, such as the memory 102 including instructions, that are executable by the processor 103 of the wearable device. For example, the non-transitory computer readable storage medium may be a ROM, a Random Access Memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

The physiological signal determination device provided in the embodiment of the present invention may be, in addition to the wearable device, a terminal device such as a mobile phone, a Personal Computer (PC), or a tablet Computer capable of acquiring a physiological signal of a human body. In the following, the method for determining the quality of a physiological signal according to the embodiment of the present invention is described by taking only the above-mentioned apparatus for determining the quality of a physiological signal as a wearable device.

Specifically, the following description of the embodiments of the present invention only takes the physiological signal as the pulse wave signal as an example, and describes the method for determining the quality of the physiological signal according to the embodiments of the present invention.

Fig. 2 is a schematic waveform diagram of a physiological signal according to an embodiment of the present invention. Fig. 2 shows a waveform of a physiological signal as a segment of a pulse wave signal in an ideal case. For the pulse wave signal in an ideal case, the signal of each period may have the same features and feature points. For example, the signal of each period in the pulse wave signal includes features such as ascending branch, main peak, dicrotic wave, descending branch, and the like, and feature points such as top point, bottom point, and the like.

Specifically, the variable on the abscissa t(s) in fig. 2 is time t, and the unit of time t is second(s); the variable on the ordinate A (mv) is the amplitude A, which is given in millivolts (mv). B is₁Dot, B₂Points and B₃The points are three bottom points, C₁Point and C₂The points are two vertices. The waveform between adjacent base points may be a signal of one cycle. Of course, the feature points for distinguishing the periods of the pulse wave signals in fig. 2 may be other feature points, such as vertices, in addition to the base points. Illustratively, bottom dot B in FIG. 2₁And the bottom point B₂The waveform in between is a signal of one period (denoted as signal period 1); the period value of signal period 1 can be recorded as T₁. Bottom point B₂And the bottom point B₃The waveform in between is the signal of another cycle (denoted as signal cycle 2); the period value of signal period 2 can be recorded as T₂. Generally, the normal period value of the pulse wave signal corresponds to the normal range of the heart rate of the human body from 40bpm to 180 bpm. Bottom point B in FIG. 2₁To F₁The waveform between the points is the main wave, F, of signal period 1₁Point to point B₂The waveform in between is the dicrotic wave of signal cycle 1.

Wherein, the bottom point B₁Which is the start of signal period 1. Bottom point B₁To the top C₁The waveform between them is the main wave ascending branch of signal period 1, and its height value can be recorded as H_1-1And the width value can be recorded as T_1-1. Vertex C₁To F₁The waveform between the points is the main wave descending branch of the signal period 1, and the height value can be recorded as H_1-2And the width value can be recorded as T_1-2。F₁Point to G₁The waveform between the points is the rising branch of the dicrotic wave of the signal cycle 1, and the height value of the rising branch can be recorded as H_1-3And the width value can be recorded as T_1-3。G₁The peak where the point is the peak of the dicrotic wave of signal cycle 1. G₁Point to bottom point B₂The wave form between the two is the dicrotic wave descending branch of the signal cycle 1, and the height value can be recorded as H_1-4And the width value can be recorded as T_1-4. Bottom point B₂The point is the end of signal period 1. Wherein, the period value T of the signal period 1₁＝T_1-1+T_1-2+T_1-3+T_1-4. Bottom point B in the signal period 1₁To the top C₁The ascending branch of the main wave of (a) can also be called a systolic wave (systolic wave) of signal cycle 1; vertex C₁To the bottom point B₂May also be referred to as the diastolic wave (diastolic wave) of signal cycle 1. Thus, the height of the contraction wave of signal period 1 is H_1-1The height of the diastolic wave of signal cycle 1 can be recorded as H_1-5。

Similarly, the characteristics and the features of the signals in other periods (e.g., signal period 2) in the pulse wave signal are similar to the signal period 1, and the detailed description of the embodiment of the present invention is omitted here.

In addition, bottom point B in signal period 2₂To E₂The area of the graph formed by the waveform of the point and the coordinate axis t (S) can be recorded as S_a；E₂Point to point B₃The area of the graph formed by the waveform and the coordinate axis t (S) can be recorded as S_b(ii) a The area of the graph formed by the waveform of the signal period 2 and the coordinate axis t (S) can be recorded as S, i.e. the bottom point B₂Point to point B₃The area of the graph formed by the waveform between the two and the coordinate axis t (S) can be recorded as S. Wherein S is S_a+S_b. Similarly, for the description of the area of the waveform of the pulse wave signal period 1 or other signal periods and the graph formed by the coordinate axes, reference may be made to the description of the signal period 2 above, and details are not repeated here.

It should be noted that, because the current wearable sensor is simple and mostly acquires a wrist signal, the pulse wave signal acquired by the current wearable sensor often cannot see features such as a dicrotic wave and feature points where the peak of the dicrotic wave is located. However, current wearable sensors can generally acquire the bottom and top points of the pulse wave signal. Similarly, the current wearable sensor can generally acquire the top points and the bottom points of other physiological signals besides the pulse wave signals, and the details are not repeated here. In the following, the method for determining the quality of a physiological signal according to the embodiment of the present invention is described by taking only the feature points of the physiological signal as the top point and the bottom point as an example.

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the following describes in detail a physiological signal quality determination method provided by the embodiments of the present invention with reference to the flowchart of the physiological signal quality determination method shown in fig. 3 in conjunction with the wearable device 10 shown in fig. 1. Referring to fig. 3, the method for determining the quality of a physiological signal according to an embodiment of the present invention may include steps S301 to S304:

s301, the wearable device collects a physiological signal, and the physiological signal is a periodic signal or a periodic-like signal.

Therein, step 301 may be performed by a sensor component 101, such as a PPG sensor, in the wearable device 10 shown in fig. 1.

For example, the sensor assembly 101 may acquire the physiological signal at a sampling frequency for a fixed time period (e.g., 3s), and the acquired physiological signal is a set of discrete samples.

The sampling frequency does not affect the implementation of the object of the present application, and the sampling frequency is not particularly limited in the embodiment of the present invention, and may be, for example, 25 to 100 Hz.

It should be noted that, after the wearable device acquires the physiological signal, the wearable device may pre-process the physiological signal, such as filtering, to acquire the pulse wave signal, as shown in fig. 4, which is the pre-processed pulse wave signal. The preprocessing process may be executed by a processor in the wearable device, or may be executed by a separate filtering component in the wearable device, where the filtering component may be a filter implemented by hardware, and the embodiment of the present invention is not limited thereto.

S302, the wearable device extracts feature points of the physiological signal, wherein the feature points comprise feature points used for indicating the period of the physiological signal.

The feature points for indicating the period of the physiological signal may be top points and bottom points of the physiological signal.

Optionally, the characteristic point may be other characteristic points in the physiological signal except for the base point and the vertex, such as a characteristic point where a peak of a dicrotic wave of the pulse wave signal is located.

The waveform between adjacent bottom points in the physiological signals (such as pulse wave physiological signals) extracted by the wearable device can be a cycle, or the waveform between adjacent top points can be a cycle; and the wearable device judges the signal quality of the physiological signal cycle by cycle, so the method provided by the embodiment of the invention may further include S303:

and S303, the wearable device divides the physiological signal into periods according to the characteristic points of the physiological signal.

Fig. 5 is a schematic diagram of waveforms of a physiological signal according to an embodiment of the present invention. Fig. 5 shows the bottom points and the top points extracted by the wearable device for the pulse wave signals shown in fig. 4 through the wearable sensor. The wearable device may then divide the cycles for the waveform as between adjacent bottom points in fig. 5 and record the signal for each cycle in time sequence, i.e. record a sequence of samples of the signal for each cycle.

Wherein, the sample sequence of the signal of the 1 st period acquired by the wearable device can be recorded as x₁＝{x_{1_1}，x_{1_2}，…x_{1_j}，…，x_{1_n}J ∈ {1, 2, … …, n }, sequence of samples x of the signal of cycle 1₁May include n samples, n being a positive integer; x is the number of_{1_j}Is a sample sequence x of the signal of the 1 st period₁The j-th sample in (a). Similarly, when the wearable device acquires a signal of any period after the 1 st period, a sample sequence of the signal of the period may also be recorded, and the sample sequence of the signal of the period may also include n samples.

It should be noted that, when the present invention is usedThe time interval between two samples in the sequence of samples of each period of the signal acquired by the wearable device may be 0.01s when the sampling frequency provided in the illustrated embodiment is 100 Hz. Combining the waveform diagrams of the physiological signals shown in fig. 3 and 5, the time T of the 1 st period of the signal shown in fig. 5₁May be 0.3s, the waveform of the 1 st period signal may include 30 points, that is, the sample sequence of the 1 st period signal may include 30 samples.

S304, the wearable device judges the signal quality of the signal of the ith period according to the similarity between the signal of the ith period and the current signal template set.

The steps 302 and 304 can be performed by the processing component 102 in the wearable device 10 shown in fig. 1.

The current signal template set comprises N signal templates, the N signal templates can be obtained according to physiological signals before the signal of the ith period, N is greater than or equal to 2, and i is greater than N. That is, the N signal templates may be N periods of signals from the 1 st period of signals to the i-1 st period of signals, and the N signal templates are acquired by the wearable device in real time. The similarity between the signal of the ith period and the current signal template set is composed of the similarity between the signal of the ith period and each signal template in the N signal templates included in the current signal template set.

Wherein, the wearable device acquires signals of every two cycles, and the signals of every two cycles have similarity. The similarity between the two periodic signals may be used to indicate the degree of similarity of the waveforms of the two periodic signals. For example, the higher the similarity between the signals of two cycles, the higher the degree of similarity of the waveforms representing the signals of the two cycles; the lower the similarity between the signals of two cycles, the lower the degree of similarity of the waveforms representing the signals of the two cycles. That is, the higher the similarity between the signal of the ith period and any signal template is, the higher the similarity between the waveform of the signal representing the ith period and the waveform of the signal template is; further, the higher the similarity between the signal of the above-mentioned i-th cycle and the current signal template set. The lower the similarity between the signal of the ith period and any signal template is, the lower the similarity between the waveform of the signal of the ith period and the waveform of the signal template is; further, the lower the similarity between the signal of the i-th cycle and the current signal template set.

Generally, each signal template of the N signal templates in the current signal template set is a good-quality physiological signal. Therefore, if the similarity between the signal of the ith period and the current signal template set is higher, the signal quality of the signal of the ith period is better; the lower the similarity between the signal of the ith period and the current set of signal templates, the worse the signal quality of the signal of the ith period.

Wherein, the wearable device can record the sample sequence of the signal of the ith period while acquiring the signal of the ith period, for example, the sample sequence of the signal of the ith period can be recorded as x_i＝{x_{i_1}，x_{i_2}，…x_{i_j}，…，x_{i_n}J ∈ {1, 2, … …, n }. At this time, the signal of the i-th cycle may be abbreviated as x_i. Wherein the sample sequence x of the signal of the ith period_iIncluding n samples, x_{i_j}Is a sample sequence x of the signal of the i-th cycle_iThe j-th sample in (a). Of course, the wearable device may have recorded a sequence of samples of the signal for each of the first i-1 cycles of the signal before the wearable device acquires the signal for the ith cycle. For example, the wearable device may record a sample sequence of the signal of the i-1 th cycle as x_i-1＝{x_{i-1_1}，x_{i-1_2}，…x_{i-1_j}，…，x_{i-1_n}Sample sequence x of the i-1 th periodic signal_i-1May also include n samples, x_{i-1_j}Is a sample sequence x of the signal of the i-1 th cycle_i-1The j-th sample in (a).

It should be noted that, in the method for judging the quality of a physiological signal provided in the embodiment of the present invention, the current signal template set for judging the quality of a physiological signal is obtained by the wearable device according to a physiological signal acquired in real time, instead of being obtained from a large amount of off-line physiological signals by using a machine learning algorithm or a priori knowledge; therefore, the current signal template provided by the embodiment of the invention is more consistent with the physiological signal acquired by the wearable device in real time, so that the accuracy of the physiological signal quality judgment result obtained according to the current signal template set is higher. Moreover, after the physiological signal is acquired, the wearable device extracts the top point and the bottom point in the physiological signal to support the wearable device to judge the quality of the physiological signal, and does not necessarily need to extract a large number of other feature points except the top point and the bottom point in the physiological signal to support the wearable device to judge the quality of the physiological signal; therefore, the calculation amount of the wearable device in the process of judging the quality of the physiological signal can be reduced, and the accuracy of the judgment result of the quality of the physiological signal can be further improved.

Specifically, in a possible implementation manner, the wearable device determines the signal quality of the signal of the ith period according to the similarity between the signal of the ith period and the current signal template set, and may determine whether the signal of the ith period is a normal physiological signal or an abnormal physiological signal. S304 in the method for determining the quality of a physiological signal according to the embodiment of the present invention may include S601-S603 or S601, S602, and S604. Exemplarily, as shown in fig. 6, a schematic flow chart of another physiological signal quality determination method according to an embodiment of the present invention is provided. In fig. 6, S304 shown in fig. 3 may include S601-S603 or S601, S602, S604:

s601, the wearable device determines a similarity result of the signal of the ith period, wherein the similarity result of the signal of the ith period is used for indicating the similarity between the signal of the ith period and the current signal template set.

The similarity between every two periodic signals in the periodic signals acquired by the wearable device can be represented by similar parameters such as correlation coefficients or mean square errors of the sample sequences of the two periodic signals. That is, the similarity between the signal of the ith period and the current signal template set may be composed of similar parameters such as correlation coefficient, mean square error, and the like between the sample sequence of the signal of the ith period and the sample sequence of one signal template. Therefore, the similarity between the signal of the ith period and the current signal template set can be represented by the similarity parameter of the signal of the ith period. As such, the similarity result of the signal of the ith period may be a similarity parameter of the signal of the ith period. The similarity parameter of the signal of the ith period may be a sum of the signal of the ith period and the similarity parameter of each of the current N signal templates, or an average value between the signal of the ith period and the similarity parameter of each of the current N signal templates.

Wherein the wearable device may record the current set of signal templates as y ═ y₁，y₂，…y_t，…，y_N-1，y_NT e {1, 2, … …, N }, this y_tIs the sample sequence of the t-th signal template in the current N signal templates. Meanwhile, the wearable device can convert the sample sequence y of the t-th signal template_tIs recorded as y_t＝{y_{t_1}，y_{t_2}，…y_{t_j}，…，y_{t_n}J ∈ {1, 2, … …, n }. Sample sequence y of the t-th signal template_tMay also include n samples, y_{t_j}For the sample sequence y of the t-th signal template_tThe j-th sample in (a). Subsequently, the wearable device may record the similar parameter of the ith periodic signal and the tth signal template of the current N signal templates as r_{i_t}. Thus, the similarity parameter r of the signal of the i-th cycle_{i_t}Can be recorded as:

or,

it should be noted that any signal template in the current signal template set y is a periodic signal, for example, the 1 st signal template y₁May be the 1 st period of signal x₁. For convenience of description, in the embodiments of the present invention, a signal template and a periodic signal corresponding to the signal template are respectively represented by different characters.

Illustratively, in the case where the similarity between two periodic signals is represented by the correlation coefficient of the sample sequence of the two periodic signals: the signal of the ith period and the current N signal templatesSimilarity parameter r of t signal templates_{i_t}Is a sample sequence x of the signal of the i-th cycle_iSample sequence y of t-th signal template in current N signal templates_tThe correlation coefficient of (2).

Specifically, the similarity parameter r between the signal of the ith period and the t-th signal template in the current N signal templates_{i_t}Can be as follows:

wherein, is the sample sequence x of the signal of the ith period_iThe average value of the n samples in the sequence is the sample sequence y of the t signal template_tAverage of n samples in (a), and

further, the similarity parameter r between the signal of the ith period and the t-th signal template in the current N signal templates is substituted into the formula (1)_{i_t}The method can also comprise the following steps:

it should be noted that, if the correlation coefficient of the sample sequence of the signals of two periods is larger, the similarity between the signals of the two periods is higher; conversely, the lower the similarity between the signals of the two cycles. Thus, if the parameter r is similar_{i_t}The larger the signal is, the higher the similarity between the signal of the ith period and the t-th signal template in the current N signal templates is; further, according to the similarity parameter r_{i_t}The obtained similarity parameter r of the signal of the ith period_iThe larger the signal quality, the better the signal quality of the signal of the i-th cycle. If the parameter r is similar_{i_t}The smaller the signal is, the lower the similarity between the signal of the ith period and the t-th signal template in the current N signal templates is; further, according to the similarity parameter r_{i_t}The obtained similarity parameter r of the signal of the ith period_iThe smaller the signal quality of the signal of the i-th period, the worse the similar result of the signal of the i-th period.

In another example, where the similarity between two periodic signals is represented by the mean square error of the sequence of samples of the two periodic signals: the similarity parameter r between the signal of the ith period and the t-th signal template in the current N signal templates_{i_t}Is a sample sequence x of the signal of the i-th cycle_iWith the t-th signal in the current N signal templatesSample sequence y of templates_tThe mean square error of (d).

Specifically, the mean square error mse between the ith periodic signal and the tth signal template in the current N signal templates_{i_t}(i.e. the similarity parameter r)_{i_t}) Can be as follows:

wherein, a_{i_j}Is a sample sequence x of the signal of the i-th cycle_iSample sequence a obtained by normalization_iThe jth sample of (a), b_{i_j}For the sample sequence y of the t-th signal template_tSample sequence b obtained by normalization_tThe j-th sample of (1).

Different from the correlation coefficient, if the mean square error of the sample sequence of the signals of two periods is smaller, the similarity between the signals of the two periods is higher; conversely, the lower the similarity between the signals of the two cycles. Thus, at the similar parameter r_{i_t}Is mean square error mse_{i_t}In the case of (3), if the similarity parameter r is_{i_t}The smaller the signal is, the higher the similarity between the signal of the ith period and the t-th signal template in the current N signal templates is; further, according to the similarity parameter r_{i_t}The obtained similarity parameter r of the signal of the ith period_iThe smaller the signal quality, the better the signal quality of the signal of the i-th period. If the parameter r is similar_{i_t}The larger the signal is, the lower the similarity between the signal of the ith period and the t-th signal template in the current N signal templates is; further, according to the similarity parameter r_{i_t}The obtained similarity parameter r of the signal of the ith period_iThe larger the signal quality, the worse the signal quality of the signal of the i-th period.

It should be noted that, in the following, the method for determining the quality of a physiological signal according to the embodiment of the present invention is described only in that the similarity between signals of two cycles is represented by the correlation coefficient of the signal sample sequences of the two cycles.

Further, the wearable device determines the signal quality of the signal in the ith period according to the similarity result of the signal in the ith period, and specifically may determine, for the wearable device, whether the signal in the ith period is a normal physiological signal or an abnormal physiological signal.

S602, the wearable device judges whether the similar result of the signal of the ith period meets a preset similar condition.

The preset similar condition may be preset by the wearable device before performing S602, and the preset similar condition may be used to indicate whether a signal of each cycle in the physiological signal acquired by the wearable device is a normal physiological signal.

It should be noted that, when the similarity between the signals of two cycles is represented by the correlation coefficient of the sample sequence of the signals of the two cycles, the range of the similarity parameter of the signals of the two cycles may be [0, 1 ]. Generally, if the similarity parameter of the signals of two periods is greater than or equal to 0.8, it indicates that the similarity between the signals of the two periods is high; if the similarity parameter of the signals of the two cycles is less than 0.8, the similarity between the signals of the two cycles is low. If the similarity parameter of the signals of the two periods is equal to 0, it indicates that there is no similarity between the signals of the two periods. If the similarity parameter of the signals of two cycles is equal to 1, the signals of the two cycles are completely similar.

If the similarity parameter of the signal in the ith period is the sum of the similarity parameter of the signal in the ith period and the similarity parameter of each signal template in the current N signal templates, the "whether the phase result of the signal in the ith period meets the preset similarity condition" may be whether the similarity parameter of the signal in the ith period is greater than or equal to 0.8 × N. If the similarity parameter of the signal in the ith period is an average value between the signal in the ith period and the similarity parameter of each signal template in the current N signal templates, the "whether the similarity result of the signal in the ith period meets the preset similarity condition" may be whether the similarity parameter of the signal in the ith period is greater than or equal to 0.8.

S603, if the similarity result of the signal in the ith period meets a preset similarity condition, the wearable device judges that the signal in the ith period is a normal physiological signal.

If the similarity result of the signal in the ith period meets the preset similarity condition, the signal quality of the signal in the ith period reaches the standard of the normal physiological signal, that is, the signal in the ith period is the normal physiological signal. Subsequently, the wearable device can mark the signal of the ith period as a normal physiological signal, and obtain human physiological information according to the signal of the ith period.

S604, if the similarity result of the signal in the ith period does not meet the preset similarity condition, the wearable device judges that the signal in the ith period is an abnormal physiological signal.

If the similarity result of the signal in the ith period does not meet the preset similarity condition, it is indicated that the signal quality of the signal in the ith period does not meet the standard of the normal physiological signal, that is, the signal in the ith period is an abnormal physiological signal. Subsequently, the wearable device may mark the ith cycle of the signal as an abnormal physiological signal and not use the ith cycle of the signal in determining the physiological information of the human body.

Further, in the event that the wearable device determines that the similar result of the signal of the ith period is better than the similar result of at least one of the current N signal templates, the wearable device may update the current set of signal templates. As after S603 or S604 described above, the method may further include S701. Fig. 7 is a schematic flowchart of another method for determining the quality of a physiological signal according to an embodiment of the present invention. In fig. 7, after S603 shown in fig. 6, the method may further include S701:

s701, the wearable device updates the current signal template set according to the signal of the ith period.

The above step 701 may be performed by the processing component 102 in the wearable device 10 shown in fig. 1.

If the signal quality of the signal of the ith period is higher than the signal quality of at least one signal template of the N signal templates in the current signal template set, the wearable device may update the current signal template set according to the signal of the ith period.

It should be noted that the current signal template set is continuously updated by the wearable device according to the physiological signals acquired in real time. The wearable device replaces the signal template with relatively poor signal quality according to the periodic signal with good signal quality to update the current signal template set, so that the signal quality of the signal template in the current signal template set is improved in real time. Therefore, the accuracy of the physiological signal quality judgment result obtained according to the current signal template set updated in real time is higher.

Generally speaking, the signal of one period used for updating the current signal template set is a normal physiological signal with better signal quality; of course, the abnormal physiological signal with better signal quality can be used. Wherein updating the current signal template by the abnormal physiological signal typically occurs during several updates after the wearable device gets the set of signal templates that are not updated.

Since the signal quality of the signal of the ith period is related to the similar result of the signal of the ith period in the embodiment of the present invention, the wearable device may update the current signal template set according to the similar result of the signal of the ith period. In another possible implementation manner, in the method for determining quality of a physiological signal provided in the embodiment of the present invention, the step S701 may specifically include steps S801 to S802. Exemplarily, as shown in fig. 8, a schematic flow chart of another physiological signal quality determination method according to an embodiment of the present invention is provided. In fig. 8, S701 in fig. 7 may specifically include S801-S802:

s801, the wearable device determines that the similarity result of the signal in the ith period is better than the similarity result of at least one signal template in the current N signal templates.

At this time, the signal quality of the signal of the ith period is higher than that of at least one signal template in the current N signal templates.

And S802, the wearable device replaces the signal template with the worst similar result in the current signal template set by using the signal of the ith period.

The wearable device replaces the signal template with the worst similar result in the current signal template set with the signal of the ith period, namely the wearable device deletes the signal template with the worst similar result from the current signal template set, and updates the current signal template set by using the signal of the ith period with the better similar result as a new signal template. The similarity result of one signal template of the N signal templates in the current signal template set can be obtained from the physiological signal before the signal of the i-th cycle.

Optionally, the above-mentioned "the wearable device uses the signal of the ith period to replace the signal template with the worst similar result in the current signal template set" may be replaced by the wearable device using the signal of the ith period to replace any signal template with the worse similar result in the current signal template set. Wherein, the similarity result of any signal template with poor similarity result is worse than that of the signal of the ith period.

It should be noted that, before the wearable device determines the signal quality of the signal of the ith period according to the similarity between the signal of the ith period and the current signal template set, the wearable device may further obtain the current signal template. Specifically, in another possible implementation manner, the method for determining the quality of a physiological signal according to the embodiment of the present invention may further include S901 before S304 or S601. Fig. 9 is a schematic flowchart of another method for determining the quality of a physiological signal according to an embodiment of the present invention. In fig. 9, S901 may be further included before S601 shown in fig. 8:

s901, the wearable device acquires a current signal template set according to a physiological signal before the signal of the ith period.

The above step 901 may be performed by the processing component 102 in the wearable device 10 shown in fig. 1.

The current signal template set may be a signal template set that has been updated by the wearable device, for example, the current signal template set may be a signal template set that has been updated by the wearable device according to the signal of the (i-1) th cycle. Alternatively, the current signal template set may be a signal template set that is not updated by the wearable device, and the current signal template set is an initial signal template set acquired by the wearable device.

Specifically, in another possible implementation, the wearable device may further acquire an initial signal template set that is not updated before acquiring the current signal template set according to the physiological signal before the signal of the ith cycle. Exemplarily, as shown in fig. 10, a schematic flow chart of another physiological signal quality determination method according to an embodiment of the present invention is provided. In fig. 10, the method shown in fig. 9 may further include, before S901:

s1001, the wearable device obtains an initial signal template set according to the signals of the first N + a periods.

The above-described step 1001 may be performed by the processing component 102 in the wearable device 10 shown in fig. 1.

Wherein, the N + a is less than i, for example, the N + a is less than or equal to i-1. A is an integer of 0 or more.

Specifically, the wearable device acquires the initial signal template set according to the signals of the first N + a cycles, which may be acquired according to the similarity of the signals of each cycle in the signals of the first N + a cycles. Thus, in another possible implementation manner, S1001 in the above method may specifically include S1101-S1102. Exemplarily, as shown in fig. 11, a schematic flow chart of another physiological signal quality determination method according to an embodiment of the present invention is provided. In fig. 11, S1001 in the method shown in fig. 10 may specifically include S1101-S1102:

s1101, the wearable device determines similarity of signals of each period in the signals of the first N + a periods.

Wherein, the similarity result of the signal of one period in the signals of the first N + a periods is composed of the similarity between the signal of the period and the signals of other periods except the signal of the period in the signals of the first N + a periods. The similarity between two periodic signals in the first N + a periodic signals may also be represented by a similarity parameter of the two periodic signals, and the similarity parameter may also be a correlation coefficient between sample sequences of the two periodic signals. Thus, a similar result of a signal of one period in the first N + a periods of the signal can be represented by a similar parameter of the signal of the period.

Referring to the above embodiments, the wearable device may also record a sample sequence of the signal for each of the first N + a cycles of the signal. For example, the wearable device may record the 2 nd cycle of the signal as x₂＝{x_{2_1}，x_{2_2}，…x_{2_j}，…，x_{2_n}J ∈ {1, 2, … …, n }; record the 3 rd periodic signal as x₃＝{x_{3_1}，x_{3_2}，…x_{3_j}，…，x_{3_n}}; record the signal of the (N + a) -1 th cycle as x_N+a-1＝{x_{N+a-1_1}，x_{N+a-1_2}，…x_{N+a-1_j}，…，x_{N+a-1_n}}; record the signal of the N + a th cycle as x_N+a＝{x_{N+a_1}，x_{N+a_2}，…x_{N+a_j}，…，x_{N+a_n}}。

For example, taking the similar result of the signal of the 1 st cycle acquired by the wearable device as an example, the similar result of the signal of one cycle in the above-mentioned signals of the first N + a cycles is illustrated: similar results for the 1 st period of the signal may be represented by similar parameters for the 1 st period of the signal. The wearable device may record a similar parameter of the 1 st cycle of the signal as r₁。

Meanwhile, the wearable device may record a similar parameter of the 1 st and 2 nd periodic signals as r_{1_2}The similarity parameter r_{1_2}May be a sample sequence x of the 1 st period signal₁With the sample sequence x of the signal of the 2 nd period₂The correlation coefficient of (2). The wearable device may record a similar parameter of the 1 st and 3 rd periodic signals as r_{1_3}The similarity parameter r_{1_3}May be a sample sequence x of the 1 st period signal₁With the sample sequence x of the signal of the 3 rd period₃The correlation coefficient of (2). The wearable device can record the similar parameter of the signal of the 1 st period and the signal of the (N + a) -1 st period as r_{1_N+a-1}The similarity parameter r_{1_N+a-1}May be a sample sequence x of the 1 st period signal₁With the sample sequence x of the signal of the (N + a) -1 st period_N+a-1The correlation coefficient of (2). The wearable device canRecording the similarity parameter of the 1 st period signal and the N + a period signal as r_{1_N+a}The similarity parameter r_{1_N+a}May be a sample sequence x of the 1 st period signal₁With the sample sequence x of the signal of the N + a th cycle_N+aThe correlation coefficient of (2).

Thus, in the case where a equals 1: similarity parameter r of 1 st period signal₁The sum of the similarity parameters of the signal of the 1 st period and the signals of other periods except the signal of the 1 st period in the signals of the first N +1 periods; or, the similar parameter of the signal of the 1 st period may also be an average value between the signal of the 1 st period and the similar parameters of the signals of the other periods except for the signal of the 1 st period in the signals of the first N +1 periods. I.e. the similar parameter of the signal of the 1 st period may be

r₁＝r_{1_2}+r_{1_3}+......+r_{1_N}+r_{1_N+1}Either the first or the second substrate is, alternatively,

in case a is not equal to 1: the similarity parameter of the 1 st period signal may be an average value between the 1 st period signal and the similarity parameters of the signals of the other periods except for the 1 st period signal in the signals of the first N + a periods. I.e. the similar parameter of the signal of the 1 st period may be

Illustratively, the similarity parameter r of the 1 st and 2 nd periodic signals_{1_2}Can be that

Wherein, is the sample sequence x of the signal of the 1 st period₁The average value of n samples in the sequence x of samples of the 2 nd period signal₂Average of n samples in (a), and

further, the similarity parameter r between the 1 st periodic signal and the 2 nd periodic signal is expressed in the formula (3)_{1_2}The method can also comprise the following steps:

similarly, the similarity parameter r for the 1 st period signal_{1_3}Similar parameter r_{1_N+a-1}Similar parameter r_{1_N+a}And other similar parameters may be referred to in the detailed description of the similar parameters r_{1_2}The detailed description of the embodiments of the present invention is omitted.

It should be noted that, after the wearable device determines the similar result of the signal of each period in the first N + a period of signals, that is, after obtaining the similar parameter of the signal of each period in the first N + a period of signals, the signal quality of the signal of each period in the first N + a period of signals can be determined. Specifically, for each cycle of the first N + a cycles of the signal: if the similarity result of the periodic signal meets the preset similarity condition, the wearable device judges that the periodic signal is a normal physiological signal; if the similarity result of the periodic signal does not meet the preset similarity condition, the wearable device judges that the periodic signal is an abnormal physiological signal.

Further, the wearable device may obtain an initial signal template set including N signal templates from the signals of the first N + a periods while determining the signal quality of the signal of each period in the signals of the first N + a periods. Specifically, in another possible implementation manner, after S1101, the method further includes S1102:

s1102, the wearable device determines the signal of the first N cycles with the best similarity result in the signals of the first N + a cycles as an initial signal template set.

The above step 1102 may be performed by the processing component 102 in the wearable device 10 shown in fig. 1.

When a is equal to 0, the N signal templates in the initial signal template set are the first N periods of signals acquired by the wearable device. When a is larger than 0, the wearable device acquires a signal of the first N + a cycles, wherein the signal of the a cycles is not a signal template in the initial signal template set.

It should be noted that, after the wearable device acquires the initial signal template set according to the first N + a cycles of signals, when receiving signals of each cycle after the N + a cycle of signals, the wearable device may further determine the current signal template set. The current signal template set may be an initial signal template set that is not updated by the wearable device, or the current signal template set may be a signal template set obtained after the wearable device updates the initial signal template set.

Accordingly, in another possible implementation manner, in the method shown in fig. 10 or fig. 11, S901 may be replaced by S1002:

s1002, the wearable device obtains a current signal template set according to the initial signal template set and physiological signals from the signal of the N + a th period to the signal of the ith period.

Accordingly, the above step 1002 may also be performed by the processing component 102 in the wearable device 10 shown in fig. 1.

Specifically, when N + a is equal to i-1, the current signal template set when the wearable device acquires the signal of the ith cycle is the initial signal template set that is not updated by the wearable device.

When the N + a is smaller than i-1, when the wearable device acquires the signal of the ith period, the current signal template set is the initial signal template set which is not updated by the wearable device; or the current signal template set is a signal template set obtained by updating the initial signal template set by the wearable device according to the physiological signal between the signal of the N + a th cycle and the signal of the i th cycle.

For example, in the case that N + a is smaller than i-1, if i-1 is equal to N + a +1 and the i-1 th period of the signal has a similar result better than that of at least one of the N signal templates included in the initial signal template set, the wearable device replaces the signal template with the i-1 th period of the signal, the signal template with the worst similar result in the initial signal template set. Subsequently, when the wearable device acquires the signal of the ith period, the current signal template set is a signal template set obtained by replacing the signal template with the worst similar result in the initial signal template set by the signal of the ith-1 period.

Optionally, the "wearable device replaces the signal template with the worst similar result in the initial signal template set with the signal of the i-1 th cycle" may be replaced with the wearable device replacing any signal template with the worse similar result in the initial signal template set with the signal of the i-1 th cycle. Wherein, the similarity result of any signal template with poor similarity result is worse than that of the signal of the i-1 th period.

For example, where N equals 5, a equals 1, i-1 equals 7, and i equals 8, the wearable device obtains an initial set of signal templates from the first 6 cycles of the signal. Assume a similarity parameter r of the 1 st period signal₁Similarity parameter r of signal equal to 0.7, 2 nd period₂Similarity parameter r of signal equal to 0.75, 3 rd period₃Similarity parameter r of signal equal to 0.8, 4 th period₄Similarity parameter r of signal equal to 0.85, 5 th period₅Similarity parameter r of signal equal to 0.85, cycle 6₆Equal to 0.9. At this time, the 5 signal templates in the initial signal template set acquired by the wearable device may be a 2 nd periodic signal, a 3 rd periodic signal, a 4 th periodic signal, a 5 th periodic signal, and a 6 th periodic signal. The wearable device may record the initial set of signal templates as y ═ x₂，x₃，x₄，x₅，x₆}。

Then, if the similarity parameter r of the signal of the 7 th period₇Equal to 0.6, the wearable device does not update the initial set of signal templates. Therefore, when the wearable device acquires the signal of the 8 th cycle, the current signal template set is the initial signal template set. That is, the 5 signal templates included in the current signal template set are still signals of 2 nd to 6 th periods; the wearable device may record the current set of signal templates as y ═ x₂，x₃，x₄，x₅，x₆}。

If the similarity parameter r of the signal of the 7 th period₇Equal to 0.95, the wearable device replaces the 2 nd cycle signal with the signal of the 7 th cycle that has the worst similar result in the initial set of signal templates. Therefore, when the wearable device acquires the signal of the 8 th cycle, the current signal template set is the signal template set obtained by updating the initial signal template set. That is, 5 signal templates included in the current signal template set are signals of periods 2 to 7; the wearable device can transmit the current signalThe template set may be recorded as y ═ x₇，x₃，x₄，x₅，x₆}。

The N signal templates in the initial signal template set acquired by the wearable device are not necessarily all normal physiological signals with good signal quality, but as long as the initial signal template set includes the N signal templates, the wearable device can be supported to execute the physiological signal quality determination method provided by the embodiment of the present invention. As the number of cycles in the physiological signal acquired by the wearable device increases, the current signal template set may be continuously updated, so that the signal quality of any signal template in the current signal template set may be improved. That is, N signal templates in the current signal template set updated by the wearable device are generally normal physiological signals with good signal quality. Therefore, the wearable device can obtain a physiological signal quality judgment result according to the current signal template set, and the accuracy is high.

It should be noted that, while the wearable device obtains the current signal template set according to the initial signal template set and the signals of the periods from the N + a-th periodic signal to the i-th periodic signal, it may also determine the similar result of the signal of each period from the N + a-th periodic signal to the i-th periodic signal. The wearable device can obtain the similar parameters of the signal in each period from the signal in the N + a th period to the signal in the ith period, so as to judge the signal quality of the signal in each period from the signal in the N + a th period to the signal in the ith period. Specifically, the signal of each period from the N + a-th periodic signal to the i-th periodic signal is: if the similarity result of the periodic signal meets the preset similarity condition, the wearable device judges that the periodic signal is a normal physiological signal; if the similarity result of the periodic signal does not meet the preset similarity condition, the wearable device judges that the periodic signal is an abnormal physiological signal.

Further, the signal of each cycle in the physiological signal acquired by the wearable device can have the same characteristic point. Therefore, the wearable device can also judge the signal quality of the physiological signal according to the characteristic value corresponding to the characteristic point of the physiological signal. Specifically, in another possible implementation manner, in the method for judging quality of a physiological signal provided in the embodiment of the present invention, S1201, S1202 and S1203 may be further included after S303. Fig. 12 is a schematic flowchart of another method for determining the quality of a physiological signal according to an embodiment of the present invention. In fig. 12, S1201, S1202, and S1203 may be further included after S303 in fig. 3:

s1201, the wearable device obtains a characteristic value of the signal of the ith period according to the characteristic point of the physiological signal.

The wearable device can acquire the period value of the signal of each period and the height value of the signal of each period according to the top point and the bottom point of the signal of each period in the physiological signal. Accordingly, the characteristic value of the signal of the i-th cycle includes: a period value of the signal of the ith period and/or a height value of the signal of the ith period. Wherein, the height value of the ith periodic signal may include a left height value of the ith periodic signal and a right height value of the ith periodic signal. Specifically, the period value of the signal of each period may be a time interval between two adjacent bottom points in the signal of the period; the height value of the signal for each period may be the amplitude of the signal for that period from the top point to each of the two bottom points.

Specifically, the wearable device may acquire the physiological signal every 3 seconds to obtain a waveform of the physiological signal. The wearable device may record the period value of the i-th period of the signal as T_i. The wearable device can obtain the period value of the signal of each period according to the number of points included in the signal of each period in the physiological signal and the time interval between two points, for example, obtain the period value T of the signal of the ith period_i. For example, in the case of a sampling frequency of 100Hz, the time interval between each two discrete intervals in the physiological signal waveform shown in FIG. 5 may be 0.01 s. At this time, the waveform of the 1 st period signal shown in fig. 5 may include 30 points, and the period value T of the 1 st period signal₁May be 0.3 s.

Wherein, the wearable device may determine the current preset threshold range according to the signal of the cycle before the signal of the ith cycle. Specifically, the wearable device may obtain the initial preset threshold range according to a mean value of feature values of all periods of signals in a group of first-obtained physiological signals. Subsequently, the wearable device may update the initial preset threshold range according to the subsequently acquired feature values of the physiological signals of all cycles to obtain the current preset cycle value range. For example, if the average value of the current period values is equal to the predetermined current period value, the predetermined current period value range may be

Similarly, the mean value of the current left-branch height value of the physiological signal acquired by the wearable device may be recorded as the mean value of the current right-branch height value may be recorded as the current preset left-branch height value range of the physiological signal acquired by the wearable device may be the current preset left-branch height value range

It should be noted that, in the embodiment of the present invention, the preset threshold range obtained by the wearable device according to the average value of the acquired feature values of the physiological signal is not unique, and may be adjusted by a related technician according to actual situations.

S1202, the wearable device determines that the characteristic value of the signal in the ith period is not within the current preset threshold range, and judges that the signal in the ith period is an abnormal physiological signal.

Specifically, the feature value of the signal in the ith period is not within the current preset threshold range, and includes: the period value of the signal of the ith period is not in the current preset period range, and/or the height value of the signal of the ith period is not in the current preset height range. For example, the period value T of the ith period signal_iWhen the value range of (b) is not in, the signal of the ith period is an abnormal physiological signal.

Accordingly, before the wearable device determines the signal quality of the signal of the ith period according to the similarity between the signal of the ith period and the current signal template set, it may also determine that the signal of the ith period is likely to be a normal physiological signal according to the characteristic value of the signal of the ith period. Specifically, before S304, the method may further include S1203:

s1203, the wearable device determines that the characteristic value of the signal in the ith period is in a current preset threshold range.

Wherein, the steps 1201 and 1203 can be executed by the processing component 102 in the wearable device 10 shown in fig. 1.

Specifically, the feature value of the signal in the ith period is in the current preset threshold range, including: the period value of the signal of the ith period is in the current preset period range and the height value of the signal of the ith period is in the current preset height range. For example, the period value T of the ith period signal_iAnd if the left branch height value of the signal of the ith period is included, and the right branch height value of the signal of the ith period is included, the signal of the ith period is possibly a normal physiological signal.

It should be noted that, in the method for determining the quality of a physiological signal provided in the embodiment of the present invention, the wearable device may determine the signal quality of the signal in each period of the physiological signal according to the period value and the height value in the feature value of the physiological signal. The wearable device can extract the characteristic points of the physiological signal to judge the signal quality of the signal in each period in the physiological signal, and the calculation amount in the process of judging the quality of the physiological signal can be reduced to a certain extent.

Further, the wearable device may determine the current preset threshold range from the physiological signal prior to the ith cycle of the signal. Specifically, the wearable device may obtain the initial preset threshold range according to a mean value of feature values of all periods of signals in a group of first-obtained physiological signals. Subsequently, the wearable device may update the initial preset threshold range according to the feature values of all cycles of the subsequently acquired physiological signals to obtain the current preset threshold range. Therefore, in another possible implementation manner, after S303, the method provided in the embodiment of the present invention may further include S1301 and S1302. Exemplarily, as shown in fig. 13, a schematic flow chart of another physiological signal quality determination method according to an embodiment of the present invention is provided. S1301 and S1302 may also be included after S303 in fig. 13:

s1301, the wearable device determines a current preset threshold range according to the physiological signal before the signal of the ith period.

Wherein, the wearable device may determine the current preset threshold range according to the signal of the cycle before the signal of the ith cycle. Specifically, the wearable device may obtain the initial preset period value range according to a mean of the period values of all periods of the first acquired group of physiological signals. For example, the wearable device may derive the initial preset period value range from the average of the period values of all periods of the signal in the physiological signal acquired in the previous 3 s. Subsequently, the wearable device may update the initial preset period value range according to the period value of the physiological signal acquired every 3s time interval before the subsequent i-th period signal, so as to obtain the current preset period value range.

Similarly, the wearable device may obtain the initial preset left-branch height value range and the initial preset left-branch height value range according to an average of the height values of all cycles of the first acquired physiological signal. Subsequently, the wearable device may update the preset left-branch height value range and the initial preset left-branch height value range according to the subsequently obtained height value of the physiological signal, so as to obtain the current preset left-branch height value range and the current preset left-branch height value range.

For example, the wearable device may obtain the initial preset left value range and the initial right value range according to the average of the height values of all cycles of the signal in the physiological signal acquired in the first 3 s. Subsequently, the wearable device may update the initial preset left count value range and the initial preset right count value range according to the height values of all periods of signals in the physiological signal acquired every 3s time interval before the subsequent signal of the ith period, so as to obtain the current preset left count value range and the current preset right count value range preset period value range.

It should be noted that the wearable device may update the current preset period value range according to the period values of all periods of signals in the physiological signal acquired at intervals of 3s after the ith period of signals, so as to obtain a new current preset period value range. Correspondingly, after S1203, the method may further include S1302:

s1302, the wearable device updates the current preset threshold range according to the characteristic value of the signal of the ith period.

The steps 1301 and 1302 can be executed by the processing component 102 in the wearable device 10 shown in fig. 1.

It should be noted that, in the method for judging quality of a physiological signal provided in the embodiment of the present invention, the current preset threshold range is continuously updated according to a physiological signal obtained in real time; the current preset threshold range is acquired by the wearable device according to all periodic signals in the acquired physiological signals, namely the current preset threshold range is obtained according to the global variable; therefore, the current preset threshold range is consistent with the physiological signals acquired by the wearable device in real time. Therefore, the method for judging the quality of the physiological signal provided by the embodiment of the invention can further improve the accuracy of the judgment result of the quality of the physiological signal.

Further, after determining the signal quality of the signal of each period in the physiological signal, the wearable device may obtain the physiological information of the human body using the determined normal physiological signal. Specifically, in another possible implementation manner, after S304, S603, or S604, the method may further include S1401. Fig. 14 is a schematic flowchart of another method for determining the quality of a physiological signal according to an embodiment of the present invention. S1401 may be further included after S304 in fig. 14:

and S1401, outputting a signal quality judgment result by the wearable device.

The above step 1401 may be performed by the processing component 102 in the wearable device 10 shown in fig. 1 through the multimedia component 104.

The signal quality judgment result comprises that the signal of the ith period is a normal physiological signal or an abnormal physiological signal, and the signal of each period before the signal of the ith period is a normal physiological signal or an abnormal physiological signal. Specifically, the wearable device may mark the normal physiological signal and the abnormal physiological signal differently, and may further output a ratio of the normal physiological signal in all periods of the acquired physiological signal, which is not limited in the embodiment of the present invention. Therefore, the judgment result of the quality of the physiological signal obtained by judging the wearable device can be visually displayed for the user or the related technical personnel, so that the user experience is better.

It should be noted that, with the method for judging the quality of a physiological signal provided in the embodiment of the present invention, the wearable device can judge the signal quality of the acquired physiological signal cycle by cycle in real time, and can more accurately distinguish whether the signal in any cycle of the physiological signal is a normal physiological signal or an abnormal physiological signal. Therefore, the wearable device can obtain more accurate human physiological information according to the normal physiological signals obtained by judgment.

The scheme provided by the embodiment of the invention is introduced mainly from the perspective of wearable equipment in the physiological signal quality judgment device. It is understood that the physiological signal quality judging device includes hardware structures and/or software modules for performing the above functions. Those of skill in the art will readily appreciate that the present invention can be implemented in hardware or a combination of hardware and computer software, in conjunction with the exemplary algorithm steps described in connection with the embodiments disclosed herein. Whether a function is performed as hardware or computer software drives hardware depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.

The embodiment of the present invention may divide the functional modules of the physiological signal quality determination device according to the above method, for example, each functional module may be divided corresponding to each function, or two or more functions may be integrated into one processing module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. It should be noted that, the division of the modules in the embodiment of the present invention is schematic, and is only a logic function division, and there may be another division manner in actual implementation.

In the case of dividing each functional module according to each function, fig. 15 shows a schematic diagram of a possible composition of the physiological signal quality determining apparatus provided in the above embodiment, and as shown in fig. 15, the physiological signal quality determining apparatus 15 may include: an acquisition module 151, an extraction module 152, a partitioning module 153, and a determination module 154. Wherein, the acquisition module 151 is configured to support the physiological signal quality determination device 15 to perform S301 in the above-described embodiment, and/or other processes for the technology described herein. The extraction module 152 is used to support the physiological signal quality determination device 15 to perform S302 in the above embodiments, and/or other processes for the techniques described herein. A partitioning module 153 for supporting the physiological signal quality determining device 15 to perform S303 in the above embodiments, and/or other processes for the techniques described herein. A determination module 154 for supporting the physiological signal quality determination device 15 to perform the processes of S304, S602, S603, S604 and S1202 in the above-described embodiments, and/or other processes for the techniques described herein.

Further, fig. 16 shows another possible composition diagram of the physiological signal quality judging device provided in the above embodiment, and as shown in fig. 16, the physiological signal quality judging device 15 may further include: a determination module 155. A determination module 155 for supporting the physiological signal quality determination device 15 to perform S601 and S1301 in the above embodiments, and/or other processes for the techniques described herein.

Further, fig. 17 shows another possible composition diagram of the physiological signal quality determining apparatus provided in the above embodiment, and as shown in fig. 17, the physiological signal quality determining apparatus 15 may further include: and an update module 156. An update module 156 for enabling the physiological signal quality determination device 15 to perform the processes S701, S801, S802, and S1302 in the above embodiments, and/or other processes for the techniques described herein.

Further, fig. 18 shows another possible composition schematic diagram of the physiological signal quality judging device provided in the above embodiment, and as shown in fig. 18, the physiological signal quality judging device 15 may further include: an acquisition module 157. An obtaining module 157, configured to support the physiological signal quality determining device 15 to perform the processes of S901, S1001, S1002, S1102, S1201, S1202 and S1203 in the foregoing embodiments, and/or other processes for the techniques described herein.

Further, fig. 19 shows another possible composition diagram of the physiological signal quality judging device provided in the above embodiment, and as shown in fig. 19, the physiological signal quality judging device 15 may further include: an output module 158. An output module 158 for supporting the physiological signal quality determination device 15 to perform S1401 in the above-described embodiments, and/or other processes for the techniques described herein.

It should be noted that all relevant contents of each step related to the above method embodiment may be referred to the functional description of the corresponding functional module, and are not described herein again.

The physiological signal quality judging device provided by the embodiment of the invention is used for executing the physiological signal quality judging method, so that the same effect as the physiological signal quality judging method can be achieved.

In the case of an integrated unit, the extraction module 152, the division module 153, the judgment module 154, the determination module 155, the update module 156, the acquisition module 157, and the like may be integrated into one processing module. The processing module may be a Processor or a controller, such as a CPU, a general purpose Processor, a Digital Signal Processor (DSP), an Application-Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), or other Programmable logic device, a transistor logic device, a hardware component, or any combination thereof. Which may implement or execute the various illustrative logical blocks, modules, and circuits described in connection with the disclosure. The processing units described above may also be combinations that perform computing functions, e.g., including one or more microprocessor combinations, DSPs and microprocessors, and the like. The storage module may be a memory. The above-mentioned acquisition module 151 may be implemented by an input device. The output module 158 may be implemented by a display.

When the processing module is a processor and the storage module is a memory, the embodiment of the present invention provides a physiological signal quality determining device 20 as shown in fig. 20. As shown in fig. 20, the physiological signal quality determination device 20 includes: processor 201, memory 222, display 203, input 204, and bus 205. Wherein the processor 201, the memory 202, the display 203 and the inputter 204 are interconnected by a bus 205. The bus 205 may be a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus 205 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one thick line is shown in FIG. 20, but this is not intended to represent only one bus or type of bus.

For example, the input device 204 may include a camera, a wearable sensor, and the like, such as the sensor assembly 101 in the wearable device 10. The display 203 may be the multimedia component 104 or the audio component 105 in the wearable device 10.

The detailed description of each module in the physiological signal quality determining device 20 and the technical effects of each module after executing the steps of the method in the foregoing embodiments of the present invention may refer to the related description in the method embodiment of the present invention, and are not repeated herein.

Through the above description of the embodiments, it is clear to those skilled in the art that, for convenience and simplicity of description, the foregoing division of the functional modules is merely used as an example, and in practical applications, the above function distribution may be completed by different functional modules according to needs, that is, the internal structure of the device may be divided into different functional modules to complete all or part of the above described functions. For the specific working processes of the system, the apparatus and the unit described above, reference may be made to the corresponding processes in the foregoing method embodiments, and details are not described here again.

In the several embodiments provided in the present application, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules or units is only one logical division, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit. The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone computer program product, may be stored in a computer readable storage medium.

When implemented in software, the technical solutions of the present application may be implemented in whole or in part in the form of a computer program product. The computer program product includes at least one instruction. When loaded and executed on a computer, cause the processes or functions described in accordance with embodiments of the invention to occur, in whole or in part. The computer may be by computer, a special purpose computer, a network of computers, or other programmable device. The instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer may transmit from one website, computer, server, or data center to another website, computer, server, or data center via a wired means such as coaxial cable, fiber optics, Digital Subscriber Line (DSL), or wireless means such as infrared, radio, microwave, etc. The computer readable medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more available media. The usable medium may be a magnetic medium such as a floppy Disk, a hard Disk, or a magnetic tape, or a semiconductor medium such as a Solid State Disk (SSD).

The above description is only an embodiment of the present application, but the scope of the present application is not limited thereto, and any changes or substitutions within the technical scope of the present disclosure should be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

A method for judging the quality of a physiological signal, comprising:

acquiring a physiological signal, wherein the physiological signal is a periodic signal or a quasi-periodic signal;

extracting feature points of the physiological signal, wherein the feature points comprise feature points used for indicating the period of the physiological signal;

dividing the physiological signal into cycles according to the characteristic points of the physiological signal;

judging the signal quality of the signal of the ith period according to the similarity between the signal of the ith period and the current signal template set; the signal template set comprises N signal templates, the N signal templates are obtained according to physiological signals before the signal of the ith period, N is greater than or equal to 2, and i is greater than N.
The method of claim 1, wherein the determining the signal quality of the signal of the ith period according to the similarity between the signal of the ith period and the signal template set comprises:

determining a similarity result of the signal of the ith period, wherein the similarity result of the signal of the ith period is used for indicating the similarity between the signal of the ith period and the current signal template set;

if the similarity result of the signal of the ith period meets a preset similarity condition, judging the signal of the ith period to be a normal physiological signal, wherein the preset similarity condition is preset;

and if the similarity result of the signal of the ith period does not meet the preset similarity condition, judging that the signal of the ith period is an abnormal physiological signal.
The method of claim 2, wherein in the event that it is determined that the similarity result of the signal of the i-th cycle is better than the similarity result of at least one of the N signal templates, the method further comprises: and updating the current signal template set according to the signal of the ith period, wherein the similar result of one signal template in the N signal templates is obtained according to the physiological signal before the signal of the ith period.
The method of claim 3, wherein the updating the current signal template set according to the signal of the ith period comprises:

and replacing the signal template with the signal of the ith period, which has the worst similar result, in the current signal template set.
The method according to any one of claims 1-4, wherein before said determining the signal quality of the signal of the i-th period based on the similarity between the signal of the i-th period and the current set of signal templates, the method further comprises: and acquiring the current signal template set according to the physiological signal before the signal of the ith period.
The method of claim 4, wherein said obtaining the current set of signal templates from a physiological signal prior to the i-th cycle of signals comprises:

acquiring an initial signal template set according to the signals of the first N + a periods;

and acquiring the current signal template set according to the initial signal template set and physiological signals from the signal of the N + a th period to the signal of the ith period.
The method of claim 6, wherein obtaining an initial set of signal templates from the first N + a cycles of the signal comprises:

determining a similarity result of the signal of each period in the signals of the first N + a periods; a is an integer greater than or equal to 0, and N + a is smaller than i; wherein the similarity result of the signal of one period in the first N + a periodic signals is composed of the similarity between the signal of the period and the signals of other periods except the signal of the period in the first N + a periodic signals;

and determining the signals of the first N periods with the best similar result in the signals of the first N + a periods as the initial signal template set.
The method of claim 7, further comprising, after said extracting feature points of the physiological signal:

acquiring a characteristic value of the signal of the ith period according to the characteristic point of the physiological signal;

determining that the characteristic value of the signal of the ith period is not within a current preset threshold range, and judging that the signal of the ith period is an abnormal physiological signal, wherein the current preset threshold range is determined according to a physiological signal before the signal of the ith period;

before the determining the signal quality of the signal of the ith period according to the similarity between the signal of the ith period and the signal template set, the method further includes: and determining that the characteristic value of the signal of the ith period is in the current preset threshold range.
The method of claim 8, wherein the characteristic value of the signal of the i-th cycle comprises: a period value of the signal of the ith period and/or a height value of the signal of the ith period;

the current preset threshold range includes: a current preset period range and/or a current preset height range; the characteristic value of the signal of the ith period is not in the current preset threshold range, and the method comprises the following steps: the period value of the signal of the ith period is not in the current preset period range, and/or the height value of the signal of the ith period is not in the current preset height range;

the characteristic value of the signal of the ith period is in the current preset threshold range, and the characteristic value comprises: the period value of the signal of the ith period is in the current preset period range, and the height value of the signal of the ith period is in the current preset height range.
The method according to claim 9, before the obtaining the feature value of the signal of the i-th cycle according to the feature point of the physiological signal, further comprising:

determining the current preset threshold range according to the signal of the period before the signal of the ith period;

after the obtaining the characteristic value of the signal of the i-th cycle according to the characteristic point of the physiological signal, the method further comprises:

and updating the current preset threshold range according to the characteristic value of the signal of the ith period.
The method according to any one of claims 1-10, further comprising:

outputting a signal quality judgment result, wherein the signal quality judgment result comprises that the signal of the ith period is a normal physiological signal or the signal of the ith period is an abnormal physiological signal, and the signal of each period before the signal of the ith period is a normal physiological signal or the signal of each period is an abnormal physiological signal;

if the similarity result of the signal in a period before the signal in the ith period meets the preset similarity condition, the signal in the period is a normal physiological signal; and if the similarity result of the signal in the period before the signal in the ith period does not meet the preset similarity condition, the signal in the period is an abnormal physiological signal.
A physiological signal quality judging device, comprising:

the acquisition module is used for acquiring physiological signals, and the physiological signals are periodic signals or quasi-periodic signals;

the extraction module is used for extracting the feature points of the physiological signal acquired by the acquisition module, wherein the feature points comprise feature points used for indicating the period of the physiological signal;

the dividing module is used for dividing the physiological signal into cycles according to the characteristic points of the physiological signal extracted by the extracting module;

the judging module is used for judging the signal quality of the signal of the ith period according to the similarity between the signal of the ith period and the current signal template set; the signal template set comprises N signal templates, the N signal templates are obtained according to physiological signals before the ith period of signals acquired by the acquisition module, N is greater than or equal to 2, and i is greater than N.
The apparatus of claim 12, further comprising:

a determining module, configured to determine a similarity result of the signal of the ith period divided by the dividing module, where the similarity result of the signal of the ith period is used to indicate a similarity between the signal of the ith period and the current signal template set;

the judgment module is specifically configured to judge that the signal of the ith cycle is a normal physiological signal if the similarity result of the signal of the ith cycle meets a preset similarity condition, where the preset similarity condition is preset; and if the similarity result of the signal of the ith period does not meet the preset similarity condition, judging that the signal of the ith period is an abnormal physiological signal.
The apparatus of claim 13, further comprising: and an updating module, configured to update the current signal template set according to the signal of the ith period when it is determined that the similarity result of the signal of the ith period is better than the similarity result of at least one signal template of the N signal templates, where the similarity of one signal template of the N signal templates is obtained according to a physiological signal before the signal of the ith period.
The apparatus according to claim 14, wherein the updating module is specifically configured to replace the signal template with the lowest similarity in the current signal template set with the signal of the ith period.
The apparatus of any one of claims 12-15, further comprising:

an obtaining module, configured to obtain the current signal template set according to a physiological signal before the signal of the ith period before the determining module determines the signal quality of the signal of the ith period according to the similarity between the signal of the ith period and the current signal template set.
The apparatus according to claim 16, wherein the obtaining module is specifically configured to obtain an initial signal template set according to the signals of the first N + a cycles; and acquiring the current signal template set according to the initial signal template set and physiological signals from the signal of the N + a th period to the signal of the ith period.
The apparatus of claim 17, wherein the determining module is further configured to determine a similarity result of the signal of each of the first N + a cycles; a is an integer greater than or equal to 0, and N + a is smaller than i; wherein the similarity result of the signal of one period in the first N + a period signals is composed of the similarity between the signal of the period and the signals of other periods except the signal of the period in the first N + a period signals;

the obtaining module is specifically configured to determine, as the initial signal template set, the first N cycles of signals with the best similarity result in the first N + a cycles of signals determined by the determining module.
The apparatus according to claim 18, wherein the obtaining module is further configured to obtain a feature value of the signal of the i-th cycle according to the feature point of the physiological signal after the extracting module extracts the feature point of the physiological signal;

the judging module is further configured to determine that the characteristic value of the signal of the ith period is not within a current preset threshold range, and judge that the signal of the ith period is an abnormal physiological signal, where the current preset threshold range is determined according to a signal of a period before the signal of the ith period; before the signal quality of the signal of the ith period is judged according to the similarity between the signal of the ith period and the signal template set, determining that the characteristic value of the signal of the ith period is in the current preset threshold range.
The apparatus of claim 19, wherein the characteristic value of the signal of the i-th cycle comprises: a period value of the signal of the ith period and/or a height value of the signal of the ith period; the current preset threshold range includes: a current preset period range and/or a current preset height range;

the characteristic value of the signal of the ith period is not in the current preset threshold range, and the method comprises the following steps: the period value of the signal of the ith period is not in the current preset period range, and/or the height value of the signal of the ith period is not in the current preset height range;

the characteristic value of the signal of the ith period is in the current preset threshold range, and the characteristic value comprises: the period value of the signal of the ith period is in the current preset period range, and the height value of the signal of the ith period is in the current preset height range.
The apparatus according to claim 20, wherein the determining module is further configured to determine the current preset threshold range according to the physiological signal before the ith cycle of signal before the obtaining module obtains the feature value of the ith cycle of signal according to the feature point of the physiological signal;

the updating module is further configured to update the current preset threshold range according to the characteristic value of the signal in the ith period after the obtaining module obtains the characteristic value of the signal in the ith period according to the characteristic point of the physiological signal.
The apparatus of any one of claims 12-21, further comprising:

an output module, configured to output a signal quality determination result, where the signal quality determination result includes that the signal in the ith cycle determined by the determination module is a normal physiological signal or the signal in the ith cycle is an abnormal physiological signal, and a signal in each cycle before the signal in the ith cycle is a normal physiological signal or the signal in each cycle is an abnormal physiological signal;

if the similarity result of the signal in a period before the signal in the ith period meets the preset similarity condition, judging that the signal in the period is a normal physiological signal; and if the similarity result of the signal in the period before the signal in the ith period does not meet the preset similarity condition, judging the signal in the period to be an abnormal physiological signal.
A physiological signal quality judging device, comprising: a processor, a memory, a display, an input device and a bus;

the memory is used for storing at least one instruction, the processor, the memory, the display and the input device are connected through the bus, when the device runs, the processor executes the at least one instruction stored in the memory, so that the device executes the physiological signal quality judging method according to any one of claims 1-11.
A computer storage medium, comprising: at least one instruction;

when the at least one instruction is executed on a computer, the computer is caused to perform the physiological signal quality determination method as defined in any one of claims 1-11.
A computer program product, comprising: at least one instruction;

when the at least one instruction is executed on a computer, the computer is caused to perform the physiological signal quality determination method as defined in any one of claims 1-11.