CN104758130B - A kind of intelligent nursing device and method based on brain-computer interface - Google Patents

Abstract

The invention discloses an intelligent nursing device based on a brain-computer interface, which is characterized in that it includes a computer, a visual stimulation screen connected to the computer, an EEG acquisition instrument, an STM32 controller connected to the computer, and a STM32 controller respectively connected to the computer. Connected electric nursing bed, alarm. The nursing device and implementation method of the present invention provide a new way of self-care for severe paralyzed patients who have been bedridden for a long time, especially for those paralyzed patients who have completely lost their motor ability and language expression ability and whose exercise space is limited to the bed. Effectively reduce the occurrence of bedsores. The invention has the advantages of precise control, stable working performance, high safety factor, universal applicability, etc., and can be used by a large number of such paralyzed patients.

Description

An intelligent nursing device and method based on brain-computer interface

technical field

The invention relates to an auxiliary device for paralyzed patients, in particular to an intelligent nursing device and method based on a brain-computer interface.

Background technique

After entering the 21st century, the number of paralyzed patients suffering from motor dysfunction and whose exercise space is limited to the bed is on the rise. Due to long-term local pressure on their backs and buttocks, it is easy to cause poor blood circulation and thus lead to the occurrence of bedsores. Therefore, they often need to change their posture from time to time to prevent bedsores with the aid of manual or other devices. Among them, some people can realize their posture changes by expressing their intentions with others or by means of auxiliary equipment (such as devices based on voice control, manual remote control switch control, etc.). But there are still a large number of people who have completely lost their motor ability and language expression ability, and they only retain the thinking ability of normal people, such as: amyotrophic lumbar sclerosis (ALS), spinocerebellar ataxia ( Paralyzed patients such as SCA) and vegetative people, these patients are often in a passive state and their postures are changed.

With people's continuous research, exploration and excavation in the field of artificial intelligence, many research results (such as timing control to change the state of the bed, artificial remote control of the working state of the bed and other smart devices) are used to assist such paralyzed patients in order to improve their health. Quality of life, but these are not changed according to the wishes of the paralyzed patients themselves. In recent years, with the rapid development of the field of neural engineering research, more and more researchers are trying to combine neural networks with smart devices in order to develop products that are more suitable for assisting the lives of such paralyzed patients. As an important branch of the field of neural engineering, the field of brain-computer interface (BCI) is developing very fast, which has aroused a wave of research in the field of brain-computer interface.

Brain-computer interface is a brain-computer communication system that does not depend on the normal output channel of the human brain. By collecting and processing and analyzing the EEG signals of the cerebral cortex, it is possible to establish a direct exchange of information between the human brain and computers or other communication devices. control channel. The human brain will produce different responses to different things, sports or cognitive activities, and thus obtain different EEG signals. After processing these signals, such as classification and feature extraction, the communication between the human brain and peripheral equipment can be realized. and control. According to the way of EEG acquisition, it can be divided into implantable acquisition and cerebral cortex contact acquisition (i.e. non-implantable); in comparison between these two acquisition methods, the EEG acquired by the implantable method has a higher noise ratio and higher accuracy , which is convenient for signal processing. However, it is necessary to implant electrodes into the human brain, which has great risks; while the amplitude of the EEG signal collected by the contact method of the cerebral cortex is small, the signal to noise is low, and it is a non-stationary signal, which is easily affected by the environment and the state of the person. The impact is relatively difficult to deal with, but there is no risk in the way. Moreover, with the continuous development of technology, people have reached a higher level of processing ability and methods for the tiny brain electrical signals of the scalp, making the application of brain-computer interface in daily life a reality. In view of these two types of situations, the EEG signal acquisition in the present invention uses the latter method, that is, non-implantable, to obtain EEG signals through electrode caps and inject conductive glue, that is, signal acquisition, and then through a series of signal processing, finally Realize the operation control of the equipment.

Although the application of brain-computer interface technology to the neurological assistance and rehabilitation of paralyzed patients is in its infancy, the application of single-brain-computer interface technology or multi-modal brain-computer interface technology to other aspects has become more common and technologically advanced. relatively mature. Such as Chinese patent (intelligent wheelchair control system based on brain-computer interface and its EEG signal processing method, publication number: CN101301244A), Chinese patent (a new type of intelligent wheelchair system based on motor imagination EEG control, publication number: CN101897640A ), which is based on a single brain-computer interface applied to the control of wheelchairs by the disabled; another example is Chinese patent (intelligent wheelchair based on multi-modal brain-computer interface, publication number: CN102309380), Chinese patent (a kind of wheelchair based on SSVEP and P300 Multi-modal EEG switch detection method, publication number: CN104090653A), these are applied in combination with various EEG signals. At present, in the research and application based on brain-computer interface, the main EEG signals used mainly include: EEG signals based on motor imagery, EEG signals based on steady-state visual evoked potential (SSVEP), Based on the P300 EEG signal (an endogenous special event-related potential (ERP) related to cognitive function, its peak value appears at about 300ms after the relevant event occurs), based on the event-related potential ( ERP) EEG signals, and chronic layer potential (SCP)-based EEG signals, etc. These EEG signals are relatively regular and scientific, and are very suitable for practical life applications.

In recent years, with the further development and promotion of brain-computer interface technology, there have been Internet browsing methods based on brain-computer interface mouse control, wireless remote control car systems, etc.; Bed auxiliary equipment, if the paralyzed patients who have been bedridden for a long time can control the nursing bed with their own brain waves according to their true wishes, so as to realize their movements such as turning over, lifting their backs, defecating, and bending their legs when they are lying on the bed. Manipulating changes in their bed position at their own will greatly improved their quality of life while also reducing nursing workload and boosting their self-care systems.

Contents of the invention

The purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art, and provide an intelligent nursing device based on a brain-computer interface.

Another object of the present invention is to provide an intelligent nursing method based on a brain-computer interface.

The purpose of the present invention is achieved through the following technical solutions:

An intelligent nursing device based on a brain-computer interface, including a computer, a visual stimulation screen connected to the computer, an EEG acquisition instrument, an STM32 controller connected to the computer, and an electric nursing bed connected to the STM32 controller, Alarm.

The EEG signal acquisition instrument includes the EEG signal acquisition instrument body, and the EEG electrode cap connected to the EEG signal acquisition instrument body selects 12 channels "FZ", "FCz", "Cz", "CPz", "Pz", "Oz", "P3", "P4", "P7", "P8", "O1", "O2" with these electrodes in their standard electrode positions .

Conductive glue is injected between the body of the EEG signal collector and the EEG electrode cap. To ensure good conductivity between the electrodes and the EEG collector.

The vertical height direction, left and right horizontal direction, front and rear direction and multiple angles of the visual stimulation screen can be adjusted. Users adjust the position of the visual stimulation screen according to their own habits and needs, so as to ensure a better vision in the states of lying flat and rolling over.

The operation function of the electric nursing bed is that the STM32 controller drives the CMOS analog switch chip MAX308 after receiving the control command to make the corresponding channel gating, so that the working state of the electric nursing bed can be controlled; , the STM32 controller will record the number of times each function has been executed by the nursing bed and feed it back to the computer through USB communication. The duration of each function operation is determined by the set parameters.

Another object of the present invention is achieved through the following technical solutions:

An intelligent nursing method based on a brain-computer interface, comprising steps in the following sequence:

S1. Adjust the position of the visual stimulation screen to ensure a better visual stimulation in the state of lying flat or rolling over;

S2. Initialize the system and set the relevant parameters of the system: the position of the function key of the wake-up controller, the time for each operation of the controller to operate the nursing bed, and the number of times of continuous operation when operating a single function of the nursing bed, which is the setting of the safe value of operation;

S3. Open the operation interface and open the visual stimulation screen, and use the human brain to produce different EEG signal characteristics for different things, sports or cognitive activities. The function keys are used to express the intention of manipulating the nursing bed, and to promote the generation of EEG signals on the cerebral cortex;

S4. Use the EEG electrode cap to collect the signal, and after the signal is filtered, amplified, and analog-to-digital converted by the EEG collector body, the data is transmitted to the computer using the Expresscard to parallel port technology;

S5. According to the state of the controller, the computer performs two different processing methods on the collected EEG signals: if the controller is in the "waiting to wake up" state, it will perform SSVEP detection and P300 potential detection; if it is in the "waiting to receive control instructions" state, only the P300 potential detection is performed; then, it is converted into a control command;

S6. After the controller receives the wake-up instruction from the computer, the controller immediately enters the state of "waiting to receive the control instruction". Through the received control instruction, it can perform a corresponding working state operation on the nursing bed or a manual call for help; control Every time the device operates a function, it will record and feed back to the computer, and then it will return to the "waiting to wake up" state;

S7. The computer judges whether each function of the nursing bed operated by the current user has reached the preset safety value in step S2; if any function exceeds the set value, then close the operation function and prompt; at this time, the user needs to execute the function once The opposite operation can make it return to normal.

In step S2, the position of the wake-up controller function key is selected among the upper left corner, upper right corner, lower left corner and lower right corner of the visual screen.

In step S2, the passage formula of the number of times that can be operated continuously when the single function of the nursing bed is operated Calculation, where N is the number of continuous operations, N takes a positive integer, and T is the nursing bed sheet with the most functions

Execution time, t is the operation time required by the user to execute a single function each time. If the single function of the nursing bed is performed for 18 seconds at most, it will reach the limit of the safety value. If a user chooses 3 seconds for each operation, then the user can execute any single function for 6 consecutive times. Of course, the user can also choose 2 times, 3 times However, if the user chooses 9s for each operation, there are only 2 options for the user to choose; similarly, if the user chooses 6s for the operation, there are only 2 and 3 options for the user.

In step S5, the SSVEP detection and P300 potential detection are specifically carried out through the following steps:

A. First, the collected EEG signal is denoised by a 1-25Hz band-pass filter, and then SSVEP detection and P300 potential detection are performed;

B. SSVEP detection: extract the EEG signal within 1s, calculate its power spectrum, and then judge the detection output of SSVEP according to whether its power spectrum exceeds the predefined threshold;

C. P300 potential detection: take the signal amplitude behind the electrode as a feature, intercept the data of the long-term window (100-600ms after stimulation) and use the support vector machine (support vector machine, SVM) model to classify the state, so that Realize the detection of P300 potential;

D. After the detection of SSVEP and P300 potential, the computer sends a wake-up command to the controller through USB communication, and at the same time starts the operation function keys on the visual stimulation screen interface to flicker, while the manual help calling function and the operation function of the nursing bed (such as lifting the back, turning left Functions such as rollover) are realized by sending corresponding control instructions from the computer to the controller after the P300 potential detection.

Compared with the prior art, the present invention has the following advantages and beneficial effects:

1. Users who are bedridden for a long time can manually call for help and control the nursing bed to start/close, turn to the left, flat to the right, turn to the right, flat to the left, lift the back, Functions such as back flat, downward bending of legs, upward leg raising, bedpan opening/closing, and quick reset nursing bed fill the vacancy in the current market and further expand the application of brain-computer interface in the field of paralyzed patients.

2. The visual stimulation screen in the present invention is designed to be adjustable in multiple directions, which can meet the needs of different users and has universal practicability.

3. The present invention can effectively reduce the occurrence of bed sores for paralyzed patients who are bedridden for a long time, greatly improve their quality of life, and reduce social burdens.

4. The present invention has the function of manually calling for help and the function of safety preset, which greatly improves its feasibility and safety.

Description of drawings

Fig. 1 is the structural representation of the intelligent nursing device based on the brain-computer interface of the present invention;

FIG. 2 is a circuit diagram of the analog switch chip Max308 of the device shown in FIG. 1 .

detailed description

The present invention will be further described in detail below in conjunction with the embodiments and the accompanying drawings, but the embodiments of the present invention are not limited thereto.

As shown in Figure 1, an intelligent nursing device based on a brain-computer interface includes a computer, a visual stimulation screen connected to the computer, an EEG acquisition instrument, an STM32 controller connected to the computer, and a computer connected to the STM32 controller. Electric nursing bed, alarm.

The EEG signal acquisition instrument includes the EEG signal acquisition instrument body, and the EEG electrode cap connected to the EEG signal acquisition instrument body selects 12 channels "FZ", "FCz", "Cz", "CPz", "Pz", "Oz", "P3", "P4", "P7", "P8", "O1", "O2" with these electrodes in their standard electrode positions .

Conductive glue is injected between the body of the EEG signal collector and the EEG electrode cap.

The vertical height direction, left and right horizontal direction, front and rear direction and multiple angles of the visual stimulation screen can be adjusted.

The operation function of the electric nursing bed is that the STM32 controller drives the CMOS analog switch chip MAX308 after receiving the control command to make the corresponding channel gating, so that the working state of the electric nursing bed can be controlled; , the STM32 controller will record the number of times each function has been executed by the nursing bed and feed it back to the computer through USB communication. The duration of each function operation is determined by the set parameters.

An intelligent nursing method based on a brain-computer interface, comprising steps in the following sequence:

S1. According to your own habits and needs, adjust the position of the visual stimulation screen to ensure that you can have a better vision in the state of lying flat or rolling over;

S2. The user puts on the electrode cap dedicated to collecting EEG, and selects 12 channels "FZ", "FCz", "Cz", "CPz", "Pz", "Oz", " P3", "P4", "P7", "P8", "O1", and "O2" and make these electrodes in their standard electrode positions, inject conductive glue to ensure good conductivity between the electrodes and the EEG acquisition instrument;

S3. System initialization and setting related parameters: the position of the function key of the wake-up controller is the upper right corner of the visual stimulation screen, the time for each operation of the controller to operate the nursing bed is 3s, and the single function of the nursing bed can be operated continuously for 6 times. The controller is in the "waiting to wake up" state, and then open the operation interface;

S4. Turn on the visual stimulation screen, and start the wake-up controller function key to flash, and the four keys (upper right corner, lower right corner, lower left corner, and upper left corner in turn) flash at 7.5Hz, 6Hz, 5.45Hz, and 6.67Hz respectively. Each of the four corners of the interface has a flashing button of the same size, but the one in the upper right corner can wake up the controller, which is set by step S3. The wake-up controller function key ("S" key for short) in the upper right corner;

S5. The user stares at the "S" key. At this time, the EEG acquisition instrument collects the EEG signals, and after denoising processing with a 1-25 Hz band-pass filter, then sends them to the computer for SSVEP detection and P300 potential detection. The SSVEP detection process is: extract the signal within 1s→calculate its power spectrum→threshold comparison; the P300 potential detection process is: intercept the data in the range of 100-600ms after stimulation→feature extraction→SVM classification; until the two After the brain signals are successfully detected, the computer sends a wake-up instruction to the controller to make it in the state of "waiting to receive control instructions"; and the operation function key on the visual stimulation screen interface is turned on to flash, and enter step S6;

S6. The function keys on the interface flash in green and black colors at random intervals of 200ms. If the user does not want to operate any function at this time, he can look at the "S" key again to make it selected, and then turn off the flickering of the operation function key. The controller is in the state of "waiting to receive control commands" and has not received any function for more than 60s. instruction, then automatically return to the "waiting to wake up" state; when the user wants to operate the nursing bed again, start from step S5. If the user wants to operate a certain function, then watch the function key icon that he expects to operate, that is, the P300 flashing key. The back function icon key in the interface. At this time, the computer only detects the P300 potential, collects and preprocesses the flashing data of each function icon key, and intercepts the data in the range of 100-600ms after stimulation for feature extraction and SVM classification to determine the function that the user wants to achieve. , after confirmation, the function icon key turns red, and the computer sends corresponding instructions to the controller through USB communication;

S7. The controller is mainly composed of STM32 as the control core. After receiving the control command, it analyzes it. If it is called for help manually, the STM32 drives the alarm to give an alarm prompt. Otherwise, drive the CMOS analog switch MAX308 chip to select its corresponding channel. By changing the resistance value of the port, as shown in Figure 2, control the two analog switch chips MAX308 through STM32, and realize the port by changing their channel gating conditions. The resistance values of COM1_6 and COM2_1 to the ground are changed, so that the operation of the nursing bed can be realized. As shown in Table 1, the resistance values of the ports COM1_6 and COM2_1 to the ground respectively correspond to the functional operation of the nursing bed. The back-up command sent in the above step S6, the controller STM32 gates the channel 3 of the U2 chip MAX308 in Figure 2, so that the resistance value from COM1_6 to the ground is 10.5KΩ;

Table 1 Resistance value corresponds to the operation function table of the nursing bed

S8. After executing the instruction, STM32 will record and feed back the number of times corresponding to each function currently executed by the nursing bed to the computer through USB communication, and the computer will judge whether the functions of the current user's operation of the nursing bed have reached the preset in step S3 safe value. If any function exceeds the set value, close the operation function and prompt. At this time, the user needs to perform a reverse operation of this function to restore it to normal.

The above-mentioned embodiment is a preferred embodiment of the present invention, but the embodiment of the present invention is not limited by the above-mentioned embodiment, and any other changes, modifications, substitutions, combinations, Simplifications should be equivalent replacement methods, and all are included in the protection scope of the present invention.

Claims (9)

1. A kind of 1. intelligent nursing device based on brain-computer interface, it is characterised in that：Including computer, it is connected respectively with computer Visual stimulus screen, electroencephalogramdata data collector, in addition to the STM32 controllers being connected with computer, and respectively with STM32 controller phases Electric care bed, alarm even；Its course of work is as follows：

   S1. the position of visual stimulus screen is adjusted, it is ensured that there can be a preferable vision thorn in the case where lying low, the state such as turn on one's side Swash；

   S2. system initialization and the relevant parameter of system is set：Wake up controller function key position, controller performs operation every time The time of care bed, and the number i.e. setting of safe operation value that can be operated continuously during operation care bed simple function；

   S3. opening operation interface and visual stimulus screen is opened, different things, motion or cognitive activities can be produced using human brain Different EEG signals features, user manipulate care bed by watching corresponding flashing function key on visual stimulus screen attentively to express It is intended to, promotes to produce EEG signals on cerebral cortex；

   S4. the signal is acquired using electrode for encephalograms cap, and this signal is filtered by electroencephalogramdata data collector body, amplified, And after analog-to-digital conversion process, turned using Expresscard and data are reached computer by vocal imitation skill；

   S5. computer carries out two kinds of different processing modes according to the state of controller to the EEG signals of collection：If controller In " waiting to wake up " state, then SSVEP detections and P300 potentiometric detections are carried out；If " control instruction to be received " state of being in, Only carry out P300 potentiometric detections；Then, control instruction is converted to；

   S6. after controller receives the wake-up instruction of computer, controller immediately enters " control instruction to be received " state, leads to The control instruction received is crossed, can once corresponding working condition operates or once manually seeks help call to care bed；Control Device processed often operates One function and can all be recorded and feed back to computer, and then it returns to " waiting to wake up " state；

   S7. computer judges whether each function of current user operation care bed has the safety for reaching and being pre-set in step S2 Value；If the functional value for exceeding setting, closes the operating function and prompts；Now, it is opposite need to perform once the function by user Operation it can be made to recover normal.
2. 2. the intelligent nursing device according to claim 1 based on brain-computer interface, it is characterised in that：Described EEG signals Acquisition Instrument includes eeg signal acquisition instrument body, and the electrode for encephalograms cap being connected with eeg signal acquisition instrument body chooses the world 12 passages " FZ ", " FCz ", " Cz ", " CPz ", " Pz ", " Oz ", " P3 ", " P4 ", " P7 ", " P8 " in 10-20 system electrodes, " O1 ", " O2 ", and these electrodes is in the electrode position of its standard.
3. 3. the intelligent nursing device according to claim 2 based on brain-computer interface, it is characterised in that：Described EEG signals Conducting resinl is injected between Acquisition Instrument body and electrode for encephalograms cap.
4. 4. the intelligent nursing device according to claim 1 based on brain-computer interface, it is characterised in that：Described visual stimulus The vertical height direction of screen, left and right horizontal direction, fore-and-aft direction and multi-angle are adjustable.
5. 5. the intelligent nursing device according to claim 1 based on brain-computer interface, it is characterised in that：Described electronic nursing Bed operating function be by STM32 controllers by receiving control instruction after, drive cmos analog switch chip MAX308, make it Corresponding passage gating, so as to control electric care bed working condition；After often having operated One function, STM32 controllers Number corresponding to each function of the current executed of care bed can be recorded and computer, each function are fed back to by USB communications The operation duration is determined by the parameter set.
6. A kind of 6. intelligent nursing method based on brain-computer interface, it is characterised in that the step of comprising following order：

   S1. the position of visual stimulus screen is adjusted, it is ensured that there can be a preferable vision thorn in the case where lying low, the state such as turn on one's side Swash；

   S2. system initialization and the relevant parameter of system is set：Wake up controller function key position, controller performs operation every time The time of care bed, and the number i.e. setting of safe operation value that can be operated continuously during operation care bed simple function；

   S3. opening operation interface and visual stimulus screen is opened, different things, motion or cognitive activities can be produced using human brain Different EEG signals features, user manipulate care bed by watching corresponding flashing function key on visual stimulus screen attentively to express It is intended to, promotes to produce EEG signals on cerebral cortex；

   S4. the signal is acquired using electrode for encephalograms cap, and this signal is filtered by electroencephalogramdata data collector body, amplified, And after analog-to-digital conversion process, turned using Expresscard and data are reached computer by vocal imitation skill；

   S5. computer carries out two kinds of different processing modes according to the state of controller to the EEG signals of collection：If controller In " waiting to wake up " state, then SSVEP detections and P300 potentiometric detections are carried out；If " control instruction to be received " state of being in, Only carry out P300 potentiometric detections；Then, control instruction is converted to；

   S6. after controller receives the wake-up instruction of computer, controller immediately enters " control instruction to be received " state, leads to The control instruction received is crossed, can once corresponding working condition operates or once manually seeks help call to care bed；Control Device processed often operates One function and can all be recorded and feed back to computer, and then it returns to " waiting to wake up " state；

   S7. computer judges whether each function of current user operation care bed has the safety for reaching and being pre-set in step S2 Value；If the functional value for exceeding setting, closes the operating function and prompts；Now, it is opposite need to perform once the function by user Operation it can be made to recover normal.
7. 7. the intelligent nursing method according to claim 6 based on brain-computer interface, it is characterised in that described in step S2 Wake-up controller function key position selected in the upper left corner of vision screen, the upper right corner, the lower left corner and the lower right corner.
8. 8. the intelligent nursing method according to claim 6 based on brain-computer interface, it is characterised in that described in step S2 Operation care bed simple function when the number that can operate continuously pass through formulaCount t to calculate, wherein N is continuous operation Number, N takes positive integer, and T is that care bed simple function at most performs the time, and t is that user performs needed for simple function every time Operating time.
9. 9. the intelligent nursing method according to claim 6 based on brain-computer interface, it is characterised in that described in step S5 SSVEP detection and P300 potentiometric detections carried out especially by following steps：

   A, first by the EEG signals collected pass through 1~25Hz bandpass filter denoising, then carry out SSVEP detection and P300 potentiometric detections；

   B, SSVEP is detected：The EEG signals in 1s are extracted, calculate its power spectrum, then whether basis exceedes to its power spectrum Predefined threshold value come judge SSVEP detection export；

   C, P300 potentiometric detections：Using the signal amplitude after electrode as feature, using the data for intercepting long-time window and using branch Hold vector machine model and carry out state classification, so as to realize the detection of P300 current potentials；

   D, computer is communicated to controller by USB and sends wake-up instruction after the detection of SSVEP and P300 current potentials, while is started and regarded Feeling stimulates the functional key flicker at screen interface, and the operating function of manually seek help vocative function and care bed is by P300 electricity Computer to controller sends corresponding control instruction to realize after the detection of position.