CN102551690B - Adaptive analysis method of human body signal - Google Patents

Abstract

The invention discloses a method for adaptively analyzing a human signal and relates to the technical field of detection of human signals. The method comprises the following steps of: 1, superposing a human signal which is acquired in real time with a standard human signal to ensure that voltage maximum in the acquired human signal and voltage maximum in the standard human signal are coincided with each other to establish an individual data model; 2, receiving one or more characteristic points which are selected from the individual data model by a user, and grouping the individual data model loaded with the characteristic points according to distribution of the characteristic points so as to determine a temporal point of the characteristic points; and 3, transmitting a feedback signal when a temporal point of an individual signal is coincided with the temporal point of the characteristic points. By the method for adaptively analyzing the human signal, a feedback point of the signal can be rapidly found by comparing the characteristic points of the individual signal with reference points of a standard signal.

Description

Adaptive analysis method of human body signal

technical field

The invention relates to the technical field of human body signal detection, in particular to an adaptive analysis method for human body signals.

Background technique

In recent years, with the increasing number of cardiovascular interventional operations, in order to be able to implement corresponding device treatment at one or more specific points in the cardiac cycle, and to promptly and accurately prompt the doctor or output signals to other operating equipment, this type of Technology is becoming more and more urgent.

Taking the big C equipment as an example, it can clearly image and position static organs in three-dimensional layers. In the existing technology, the front and side images of the patient are imaged twice, and the data of the two images are matrix transformed to form a three-dimensional image. However, for dynamic organs such as the heart, using this method to image not only cannot obtain valuable images, but Patient irradiated multiple times due to poor imaging.

The present invention can perform frontal or side irradiation imaging on the patient at designated feature points of different cardiac cycles of the patient. Since the cardiac time directions corresponding to the feature points are consistent, the imaging effect is better, and the number of irradiations of the patient is reduced at the same time.

Contents of the invention

(1) Technical problems to be solved

The technical problem to be solved by the present invention is: how to output the feedback signal at one or some specific points in the cardiac cycle.

(2) Technical solution

In order to solve the above-mentioned technical problems, the present invention provides a human body signal adaptive analysis method, comprising the following steps:

S1: Superimpose the real-time collected human body signal with the standard human body signal, so that the time point of the maximum voltage value in the collected human body signal and the standard human body signal is consistent, so as to establish an individual data model;

S2: Receive one or more feature points selected by the user on the individual data model, and group the individual data model after loading the feature points according to the distribution of feature points to determine the time point where the feature points are located;

S3: Send a feedback signal when the time point of the individual signal coincides with the time point of the feature point.

Wherein, the specific method of grouping in the step S2 is: after grouping the individual data models, according to the division of the reference points, the feature point specified by the user must be at a certain position between the two reference points, so between the two points with Time unit integration can determine the time point where the feature point is located.

Wherein, if the heart rate of the collected human body signal is an unlearned heart rate, the grouping in step S2 also includes: learning the individual data model to adjust the time points at which feature points appear under different heart rate conditions .

Wherein, the process of learning individual signals includes:

For a period of time, collect human body signals and calculate the heart rate;

The ECG signal based on the current heart rate is superimposed on the standard human body signal, so that the collected signal carries the reference point;

According to the feature points specified by the user, the time point where the feature points are located under the current heart rate is calculated after Fourier expansion.

Wherein, before sending the feedback signal in step S3, it also includes: when the time domain of the individual signal coincides with the time point of the feature point, adjust the feedback signal to advance or delay the specified number of milliseconds, and send the feedback signal according to the adjusted time point.

Wherein, after the step S3, it also includes recording individual signal sources, user-specified feature points and feedback signal points.

Wherein, the step S1 also includes the process of preprocessing the human body signal detected in real time before the comparison:

Extract the signal from the noise, perform data extraction and compression, perform smoothing, baseline correction and digital filtering.

Wherein, the human body signal includes: ECG and/or invasive blood pressure signal.

(3) Beneficial effects

The human body signal adaptive analysis method of the present invention can quickly find the feedback point of the signal by comparing the characteristic point of the individual signal with the standard signal reference point. When using medical equipment for imaging, the patient can be irradiated from the front or side at the specified feature points of different cardiac cycles of the patient. Since the cardiac time directions corresponding to the feature points are consistent, the imaging effect is better, and the number of irradiations of patients is reduced at the same time. .

Description of drawings

Fig. 1 is a flow chart of a method for adaptive analysis of human body signals according to an embodiment of the present invention;

Figure 2 is a comparison chart of ECG and aortic blood pressure at the same time using the above method.

Detailed ways

The specific implementation manners of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

As shown in Figure 1, the human body signal adaptive analysis method of the present invention comprises:

Step S101, superimpose the human body signal collected in real time with the standard human body signal, so that the time point of the maximum voltage in the collected human body signal and the standard human body signal is consistent, so as to establish an individual data model, and output each Signal waveforms, human body signals usually refer to ECG and/or invasive blood pressure signals (these two signals can be analyzed separately or simultaneously). The standard ECG and invasive blood pressure signal data that comes with the device are each composed of 2000 reference point connections. The standard ECG signal is a sinus heart rate of 80 beats per minute, and the output beat rate of the invasive blood pressure is the same. Since the real-time detected ECG or invasive blood pressure signal is not necessarily within this range, the standard signal is divided by the interval movement method. The data is enlarged or reduced to fit the collected signal data. From the output waveform, the time coordinates of the voltage maximum value in the collected human body signal and the standard human body signal are the same. A set of data models formed by overlapping standard signals and collected human body signals is called individual data models. The individual data model truly reflects the human body signal of the collected person, and at the same time uses the reference point to segment the collected ECG and invasive blood pressure signals into regions. The screen outputs the amplified ECG and invasive blood pressure signal waveforms.

The standard human body signal with ECG and invasive blood pressure in the acquisition equipment is marked with several reference points for the standard signal according to the different characteristics of ECG or invasive blood pressure. Since the human body signal is very complex and mixed with a lot of noise, preferably, before the comparison, it also includes: extracting the real-time signal from the noise, performing data extraction and compression, performing smoothing processing, baseline correction, digital filtering and related operations.

The above-mentioned devices for collecting real-time ECG and invasive blood pressure signals include:

The electrocardiographic signal pickup device uses the Zn-Cu electrode as the skin electrode to measure the potential generated by the electrical activity of the heart.

The invasive blood pressure signal is a piezoelectric-electric conversion device, and the invasive blood pressure sensor is used as the blood pressure sensor.

Weak signal amplification device: The ECG signal amplifier adopts three-stage amplification, the pre-stage input noise is 50 microvolts, high input impedance, common mode rejection ratio above 80db, and frequency response of 0.2-200Hz. The invasive pressure signal amplifier collects two-stage differential amplification, and has corresponding temperature compensation, and the amplification factor is adjustable from 500 to 3000.

Low-pass, high-pass and band-stop filters for noise reduction.

Among them, the frequency range of the main adult ECG signal is 1-100Hz. In order to suppress noise and facilitate the work of the subsequent stage, after the preamplifier, design and band-stop, low-pass and high-pass filters, and the center frequency of the band-stop filter 50Hz, the notch depth is 40db, and the Q value is 0.75; the cutoff frequency of the low-pass filter is 150Hz, and the frequency above 150Hz has 6db/octave attenuation; the cutoff frequency of the high-pass filter is 1Hz, and the low-end frequency has 5db/time frequency attenuation.

The invasive pressure signal is encapsulated by the invasive pressure signal sensor, and can be used after differential isolation.

Acquisition and hold and analog-to-digital conversion device for the collected signal;

Due to the high acquisition frequency, there is an acquisition and hold circuit before the analog/digital conversion, and CPLD is used to control multiple gates. The analog/digital conversion chip is MAX1168, and the real-time 8-way 16-bit sampling is fixed at 100KHz.

Auxiliary equipment includes: a detection device for electrode shedding, a device for realizing man-machine dialogue, and a feedback signal sending port.

In step S102, the receiving user selects one or more feature points on the individual data model, and groups the individual data models after the feature points are loaded. Specifically, the user selects one or more feature points on the output individual data model through the human-computer interaction device. After the Fourier expansion of the individual data model, according to the division of the reference point, the feature point specified by the user must be at a certain position between the two reference points, so the integration between the two points in time units can determine the location of the feature point point in time.

In addition to the Fourier transform, grouping can also be performed by the third-order derivative method and the interval moving method.

Step S103, send a feedback signal when the time point of the individual signal coincides with the time point of the feature point, and adjust the feedback signal to advance or delay the specified number of milliseconds, and send the feedback signal according to the adjusted time point. Record individual signal sources, user-specified feature points and feedback signal points after sending.

Before steps S102 and S103, if the heart rate of the collected human body signal is an unlearned heart rate (the heart rate may be different in different time periods, and the time points at which the adjustment feature points appear are also different under different heart rate conditions, for example: when the heart rate is 80 beats The time point at which the feature point appears is adjusted at the time of / minute, and when the heart rate of 80 beats / minute is collected next time, there is no need to learn, otherwise the following learning process is carried out), then after grouping in the step S2, it also includes: The signal is learned to adjust the timing of feature points under different heart rate conditions. The specific learning process is as follows:

For a period of time, collect human body signals and calculate the heart rate;

The ECG signal based on the current heart rate is superimposed on the standard human body signal, so that the collected signal carries the reference point;

According to the feature points specified by the user, the time point where the feature points are located under the current heart rate is calculated after Fourier expansion.

When imaging with medical equipment, the above method can be used to perform frontal or side irradiation imaging on the patient at the specified feature points of different cardiac cycles of the patient. Since the cardiac rhythms corresponding to the feature points are in the same direction, the imaging effect is better, and at the same time, the number of patients is reduced. The number of irradiations.

As shown in Figure 2, it is assumed that the feature points specified by the user are in two states: the aortic valve is open and the aortic valve is closed.

According to the self-adaptive feature of the present invention, human body ECG and invasive blood pressure signals are collected first, and superimposed with standard human body ECG and invasive blood pressure signals, so that the collected human body signal and the standard human body signal have the maximum voltage value The timing is the same. The standard signal is composed of several reference points connected, and has been clinically verified. This reference point is relatively consistent with the working state of the heart.

After the signal learning model was established in the previous step, the new ECG signals collected later are compared with the learning model. The comparison during this period is mainly to correct the heart rate, signal voltage value, and convert the signal trend to correct the learning model.

Combining the revised learning model with the user-specified feature points, the peaks of the R wave and S wave in the legend, the apex of the T wave, and the rising inflection point of the notch of the blood pressure signal are the reference points.

Superimpose the above reference points to the corrected learning model to form an individual signal data model.

According to the requirements, the first user-designated point is between the R wave and the S wave. When the new corrected learning model data is processed by a separate trend analysis algorithm, the software can determine whether the data is close to or reaches the user-designated point. In this way, the sending of the feedback signal is controlled.

The judgment method of the second user designated point is the same as that of the first designated point.

The present invention can analyze the ECG signal and the invasive blood pressure signal separately, and also can analyze both simultaneously.

The above embodiments are only used to illustrate the present invention, but not to limit the present invention. Those of ordinary skill in the relevant technical field can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, all Equivalent technical solutions also belong to the category of the present invention, and the scope of patent protection of the present invention should be defined by the claims.

Claims (7)

1. a human body signal adaptive analysis method is characterized in that, may further comprise the steps:

S1: the human body signal that will collect in real time and the stack of the human body signal of standard, the time point of the appearance voltage max in the human body signal of the feasible human body signal that collects and standard is consistent, to set up the individual data items model, described human body signal comprises: electrocardio and/or invasive blood pressure signal;

S2: receive one or more characteristic points that the user selects at the individual data items model, the individual data items model after the loading characteristic point is divided into groups to determine the time point at characteristic point place by the distribution of characteristic point;

S3: when individual signal time point overlaps with the characteristic point time point, send feedback signal.

2. human body signal adaptive analysis method as claimed in claim 1, it is characterized in that, the concrete mode of dividing into groups among the described step S2 is: after the grouping of individual data items model, according to cutting apart of datum mark, user's characteristic specified point one fixes on certain position between two datum marks, so with the unit of time integration, can determine the time point at characteristic point place at point-to-point transmission.

3. human body signal adaptive analysis method as claimed in claim 1, it is characterized in that, if the heart rate of the human body signal that collects is the heart rate of not learning, then also comprise after the grouping among the described step S2: described individual data items model is learnt, under different heart rate situations, to adjust the time point that characteristic point occurs.

4. human body signal adaptive analysis method as claimed in claim 3 is characterized in that, the process that described individual signal is learnt comprises:

In a period of time, gather human body signal and calculate heart rate;

Superpose according to the electrocardiosignal of current heart rate and the human body signal of standard, make the signal that collects carry datum mark;

According to user's specific characteristic point, calculate the time point at characteristic point place under current heart rate after the Fourier expansion.

5. human body signal adaptive analysis method as claimed in claim 1, it is characterized in that, also comprise before sending feedback signal among the step S3: when individual signal time point overlaps with the characteristic point time point, adjust feedback signal in advance or delay an appointment millisecond number, send feedback signal by the time point after adjusting.

6. human body signal adaptive analysis method as claimed in claim 1 is characterized in that, also comprises recording individual signal source, user's specific characteristic point and feedback signal point after the described step S3.

7. human body signal adaptive analysis method as claimed in claim 1 is characterized in that, also comprises the pretreated process of the human body signal that real-time detects before contrast among the described step S1:

Signal is extracted from noise, carry out the data pick-up compression, carry out smoothing processing, baseline correction and digital filtering.

Images