Hand fine movement function rehabilitation evaluation method and system

By collecting and processing the hand and brain signals of stroke patients and extracting characteristic parameters, the in-depth evaluation of the fine motor function of hand in stroke patients is achieved, solving the inefficiency and inaccuracy of existing evaluation methods, and providing personality Rehabilitation plan.

CN118924283AInactive Publication Date: 2024-11-12NAT REHABILITATION ASSISTIVE DEVICES RES CENT
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
CN202410856685.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-28
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The existing rehabilitation evaluation methods for hand motor function in stroke patients are complex in operation, low in accuracy, and inefficient, and cannot deeply and accurately evaluate the rehabilitation status and rehabilitation progress of the fine motor function of the patients in their hands.

Method used

By collecting the tactile force change signals and multi-channel cerebral blood oxygen change signals when the subject is tested performing hand test actions, performing adaptive sliding average filtering and noise removal, using the self-attention mechanism to select more important channels, extracting Tactile force and cerebral blood oxygen characteristic parameters to achieve evaluation of finger force and brain function.

Benefits of technology

Accurate assessment of the fine motor function of hand in stroke patients is achieved, and scientific basis is provided to adjust the rehabilitation plan and improve the pertinence and personalization of the rehabilitation plan.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118924283A_ABST
    Figure CN118924283A_ABST
Patent Text Reader

Abstract

The invention provides a hand fine movement function rehabilitation evaluation method and system. The method comprises the steps that touch force change signals and multi-channel brain blood oxygen change signals generated when a tested person executes hand testing actions are collected; adaptive moving average filtering processing is carried out on the touch force change signals, and noise and motion artifact removing processing is carried out on the multi-channel brain blood oxygen change signals; performing feature extraction on the processed touch force change signal to obtain a touch force feature parameter; partial channels with higher attention importance scores in the processed brain blood oxygen change signals are selected through a self-attention mechanism; feature extraction is conducted on the brain blood oxygen change signals of the partial channels with the higher attention importance scores, and brain blood oxygen feature parameters are obtained; and obtaining a finger force evaluation result based on the tactile force characteristic parameters, and obtaining a brain function evaluation result based on the brain blood oxygen characteristic parameters, thereby realizing accurate rehabilitation evaluation of the hand fine movement function.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the fields of stroke rehabilitation assessment and computer technology, and in particular to a hand fine motor function rehabilitation assessment method and system. Background Art

[0002] Stroke is a common cerebrovascular disease, usually caused by the rupture or blockage of cerebral blood vessels, resulting in damage to brain tissue. This condition often causes patients to have varying degrees of hand motor dysfunction. Therefore, rehabilitation assessment of hand motor function is the core content of stroke recovery. Hand function impairment can affect multiple aspects of movement, such as weakened strength, decreased finger control ability, and abnormal finger strength control. Stroke patients usually lose the perception of limb strength, which makes effective rehabilitation treatment particularly important, and this requires adjustment of treatment plans based on continuous functional assessment. Although conventional motor function assessment scales can provide basic functional assessments, the introduction of sensor technology has significantly improved the accuracy of motor function assessments. However, these methods are still not enough to gain a deep understanding of the rehabilitation progress of patients' hand motor function.

[0003] With the country's emphasis on the construction of medical undertakings and the development of rehabilitation medicine and artificial intelligence, the methods of post-stroke rehabilitation training and evaluation have been improved. The rehabilitation of motor function undergoes a process of assessment-rehabilitation treatment-reassessment. Among them, the classic assessment of the rehabilitation status of hand function is mostly conducted by doctors through one-on-one observation and subjective assessment, with the addition of some medical imaging assistance. These methods are inefficient and overly dependent on manual labor, and medical imaging requires high costs and complex operating techniques, which puts certain pressure on patients economically and psychologically.

[0004] At present, hand motor function rehabilitation assessment of stroke patients is an important part of stroke rehabilitation assessment, but the existing hand function assessment methods are complicated to operate, have low assessment accuracy, take up a long time of patients and are inefficient, and cannot deeply and accurately assess the rehabilitation status and progress of patients' fine motor function of the hand. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides a method and system for evaluating the rehabilitation of fine motor function of the hand to eliminate or improve one or more defects existing in the prior art.

[0006] One aspect of the present invention provides a method for evaluating the rehabilitation of fine motor function of hands, the method comprising:

[0007] Collect the tactile force change signal and multi-channel brain blood oxygen change signal when the subject performs the hand test action;

[0008] Performing adaptive sliding average filtering on the tactile force change signal, and removing noise and motion artifacts from the multi-channel cerebral blood oxygen change signal;

[0009] Extracting features of the processed tactile force change signal to obtain tactile force feature parameters;

[0010] The self-attention mechanism is used to select some channels with higher attention importance scores in the processed brain blood oxygen change signal;

[0011] Feature extraction is performed on the cerebral blood oxygen change signals of some channels with higher attention importance scores to obtain cerebral blood oxygen feature parameters;

[0012] A finger force evaluation result is obtained based on the tactile force characteristic parameter, and a brain function evaluation result is obtained based on the cerebral blood oxygen characteristic parameter, thereby realizing a rehabilitation evaluation of the fine motor function of the hand.

[0013] In some embodiments of the present invention, selecting some channels with higher attention importance scores in the processed cerebral blood oxygen change signal by the self-attention mechanism includes:

[0014] Calculate the correlation between the processed cerebral blood oxygen change signals of every two channels in the multi-channels based on the Pearson correlation coefficient, and construct an adjacency matrix corresponding to all channels in the cerebral blood oxygen change signal based on the correlation, wherein if the correlation is greater than the Pearson correlation coefficient threshold, the element value at the corresponding channel position in the adjacency matrix is ​​1, otherwise the element value at the corresponding channel position in the adjacency matrix is ​​0, and the element values ​​on the main diagonal of the adjacency matrix are all 0;

[0015] Constructing a graph based on the adjacency matrix, the graph including nodes and edges for indicating connection relationships between nodes, wherein the processed cerebral blood oxygen change signals of the corresponding channels whose element values ​​in the adjacency matrix are 1 are used as nodes of the graph, and the correlations between the processed cerebral blood oxygen change signals of the corresponding channels are used as edges of the graph;

[0016] Inputting the graph into a graph convolutional network of a self-attention mechanism, so that the graph convolutional network of the self-attention mechanism outputs a plurality of attention weights corresponding to each node, and obtaining an attention importance score of each node in the graph based on the plurality of attention weights corresponding to each node, wherein the attention weight is used to indicate the importance of each node to each of its neighboring nodes;

[0017] Sort the attention importance scores of each node in order of high to low, and select some channels with higher attention importance scores.

[0018] In some embodiments of the present invention, the tactile force characteristic parameters include maximum grip force, force duration, force change rate and force stability; the cerebral blood oxygen characteristic parameters include cerebral blood oxygen activation level and clustering coefficient and global efficiency of brain functional network connection level.

[0019] In some embodiments of the present invention, obtaining a finger force evaluation result based on the tactile force characteristic parameter includes:

[0020] By comparing and analyzing the tactile force characteristic parameters with the corresponding health indicators of healthy people of the same gender and age group as the test subject, and comparing and analyzing the tactile force characteristic parameters obtained at different rehabilitation stages of the test subject, the finger force evaluation result of the test subject is obtained; the finger force evaluation result includes the grade assessment of finger force control ability and finger force sensitivity, the finger force control ability includes the maximum gripping force and the duration of the force, and the finger force sensitivity includes the rate of change of the force and the stability of the force.

[0021] In some embodiments of the present invention, obtaining a brain function assessment result based on the cerebral blood oxygen characteristic parameter includes:

[0022] By comparing and analyzing the clustering coefficient and global efficiency of the cerebral blood oxygen activation level and the brain functional network connectivity level with the corresponding health indicators of healthy people of the same gender and age group as the test subject, and comparing and analyzing the clustering coefficient and global efficiency of the cerebral blood oxygen activation level and the brain functional network connectivity level obtained at different rehabilitation stages of the test subject, a grade assessment of the cerebral blood oxygen activation level and the brain functional network connectivity level of the test subject is obtained.

[0023] In some embodiments of the present invention, the hand test movements include part or all of the following: thumb-to-finger, fingertip-to-finger, three-finger pinch, cylinder grip, side pinch and fist.

[0024] In some embodiments of the present invention, the tactile force change signal is obtained by collecting the tactile force change signals of the thumb, index finger, middle finger, ring finger and little finger through a capacitive pressure sensor, and the multi-channel cerebral blood oxygen change signal is obtained by collecting through a multi-channel near-infrared brain functional imaging device covering the prefrontal lobe, premotor area, supplementary motor area, primary motor area, sensorimotor area and occipital lobe.

[0025] In some embodiments of the present invention, the method further comprises:

[0026] A rehabilitation assessment database for the test subject is established, the finger force assessment results and the brain function assessment results are stored in the rehabilitation assessment database in real time, and the rehabilitation assessment results of the test subject's hand fine motor function are displayed in a report.

[0027] Another aspect of the present invention provides a hand fine motor function rehabilitation assessment system, the system comprising: a processor, a memory and computer instructions stored in the memory, the processor is used to execute the computer instructions, and when the computer instructions are executed, the system implements the steps of the aforementioned method.

[0028] Another aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which implements the steps of the aforementioned method when executed by a processor.

[0029] Another aspect of the present invention provides a computer program product, comprising computer instructions, which implement the steps of the above method when executed by a processor.

[0030] The hand fine motor function rehabilitation assessment method and system of the present invention can comprehensively and deeply assess the rehabilitation of the hand fine motor function of stroke patients from multiple angles from two levels: physiological and neural feedback. By simultaneously collecting the tactile force signals and multi-channel brain blood oxygen change signals generated by the patient when performing fine hand movements, the signal data is processed by denoising, channel selection, and feature extraction, and the finger force physiological parameter index and brain function neural parameter index used to assess the rehabilitation of stroke patients are obtained. By deeply studying the brain function response and finger force physiological feedback information during fine hand movements, an accurate assessment of the hand fine motor function of stroke patients is achieved. This is conducive to formulating or adjusting the rehabilitation plan according to the patient's specific rehabilitation assessment situation, providing a scientific basis for it, and making the rehabilitation plan more targeted and personalized.

[0031] Additional advantages, purposes, and features of the present invention will be described in part in the following description, and will become apparent to those skilled in the art after studying the following, or may be learned from the practice of the present invention. The purposes and other advantages of the present invention may be achieved and obtained by the structures specifically indicated in the specification and the accompanying drawings.

[0032] Those skilled in the art will appreciate that the objectives and advantages that can be achieved with the present invention are not limited to the above specific description, and the above and other objectives that can be achieved by the present invention will be more clearly understood from the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The drawings described herein are used to provide a further understanding of the present invention, constitute a part of this application, and do not constitute a limitation of the present invention.

[0034] Figure 1 It is a flow chart of a method for evaluating the rehabilitation of hand fine motor function in one embodiment of the present invention;

[0035] Figure 2A schematic diagram of a test subject's personal information interface displayed in an embodiment of the present invention;

[0036] Figure 3 Schematic diagram of the test subject's hand fine motor function rehabilitation assessment results in one embodiment of the present invention. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments and the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0038] It should also be noted that, in order to avoid obscuring the present invention due to unnecessary details, only structures and / or processing steps closely related to the solutions according to the present invention are shown in the accompanying drawings, while other details that are not closely related to the present invention are omitted.

[0039] It should be emphasized that the term “include / comprises” when used herein refers to the presence of features, elements, steps or components, but does not exclude the presence or addition of one or more other features, elements, steps or components.

[0040] It should also be noted that, unless otherwise specified, the term “connection” herein may refer not only to a direct connection but also to an indirect connection involving an intermediate.

[0041] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the accompanying drawings, the same reference numerals represent the same or similar components, or the same or similar steps.

[0042] With regard to the rehabilitation assessment of hand motor function in stroke patients, the existing assessment methods cannot simultaneously assess the rehabilitation progress of the patient's hand motor function through overall feedback from multiple angles of the hand and the brain, nor can they achieve an in-depth and accurate assessment of the rehabilitation of the patient's hand fine motor function, thereby failing to accurately assess and understand the rehabilitation of stroke patients. To this end, the embodiments of the present invention provide a method and system for assessing the rehabilitation of hand fine motor function in stroke patients, which can meet the rehabilitation assessment needs of the field of rehabilitation medicine for the rehabilitation of hand fine motor function in stroke patients, achieve accurate rehabilitation assessment and timely understand the rehabilitation progress, thereby reflecting the overall rehabilitation of stroke patients.

[0043] Figure 1 FIG. 1 is a flow chart of a method for evaluating the rehabilitation of fine motor function of the hand according to an embodiment of the present invention. Figure 1 As shown, a method for evaluating hand fine motor function rehabilitation provided by one embodiment of the present invention comprises the following steps:

[0044] Step S110, collecting tactile force change signals and multi-channel brain blood oxygen change signals when the subject performs fine hand movements, wherein the fine hand movements include part or all of the following: thumb-to-finger, fingertip-to-finger, three-finger pinch, cylindrical grip, side pinch and fist.

[0045] Specifically, when the subject (stroke patient) performs fine hand movements, the wearable finger-cuff capacitive pressure sensor and the head-cap multi-channel near-infrared brain functional imaging device can synchronously collect the signals of the subject's tactile force and brain blood oxygen change over time, that is, the tactile force change signal and the brain blood oxygen change signal. Among them, the finger-cuff capacitive pressure sensor adopts an array electrode arrangement capacitive pressure sensor, which mainly closely adheres to the fingertip area of ​​the subject's finger to accurately collect the tactile force change signal of a single finger, part of the fingers or all the fingers of the thumb, index finger, middle finger, ring finger and little finger when the subject performs fine hand movements such as thumb-to-finger, fingertip-to-finger, three-finger pinch, cylindrical grip, side pinch and fist.

[0046] The multi-channel near-infrared brain function imaging device uses a 63-channel near-infrared brain function imaging device that can cover the frontal lobe, premotor area, supplementary motor area, primary motor area, sensorimotor area, and occipital lobe. It mainly collects 63 channels of brain blood oxygen change signals when the test subject performs fine hand movements such as thumb-to-finger, fingertip-to-finger, three-finger pinch, cylindrical grip, side pinch, and fist by surrounding the test subject's head with a head cap. The collection process of brain blood oxygen information is a multi-block paradigm. Different blocks are selected according to different test subjects. Each block includes 30s of tasks and 30s of rest, that is, the collected multi-channel brain blood oxygen change signals include task-state brain blood oxygen change signals and resting-state brain blood oxygen change signals. The above collection process is simple, fast and efficient. The test subject only needs to wear finger sleeves and head caps to start collecting the corresponding signals. The collection time is only about six to ten minutes.

[0047] In one embodiment of the present invention, before collecting the tactile force signal and the brain blood oxygen signal in step S110, the method can first register and log in to the testee's personal account, and support the addition, modification and deletion of the testee's basic information under the personal account, so as to establish a personal rehabilitation assessment file. After the personal file is established, the signal collection can be started. The basic information may include the testee's name, gender, age, height, weight, stroke type, hemiplegic side and assessment date, etc. Figure 2 shown.

[0048] Step S120 , performing adaptive sliding average filtering on the tactile force change signal, and removing noise and motion artifacts from the multi-channel cerebral blood oxygen change signal.

[0049] Specifically, for the collected tactile force change signal, an adaptive sliding average filter is used for adaptive sliding average filtering processing, which can automatically adjust the influence of each data point according to the changing characteristics of the signal, so that it can retain the main characteristics of the signal while minimizing the influence of random noise. The specific formula for filtering processing is: F(t) = α·f(t)+(1-α)·F(t-1), where f(t) is the original tactile force signal collected that changes with time t; α is the smoothing factor, which can be automatically adjusted according to the variability of the signal, thereby improving the filtering effect. Therefore, through adaptive sliding average filtering processing, noise signals can be eliminated and tactile force signal data can be smoothed, and error fluctuations caused by extreme values ​​can be reduced to obtain a relatively pure tactile force change signal.

[0050] For the processing of the collected multi-channel cerebral blood oxygen change signals, this embodiment uses a fourth-order Butterworth bandpass filter to remove part of the physiological noise and instrument noise of the collected original signal, and detects and removes the motion artifacts of the original cerebral blood oxygen change signal through the moving standard deviation method to obtain a relatively pure oxygenated hemoglobin HbO2 concentration change signal. The Butterworth bandpass filter has a smooth transition between the passband and the stopband, which can effectively remove high-frequency noise and other frequency interference. The moving standard deviation method specifically sets the threshold through a sliding window, and calibrates the part of the signal that exceeds the threshold as containing motion artifacts, and the part that does not exceed the threshold is considered to not contain motion artifacts, and then the detected motion artifacts are removed. The HbO2 concentration change signal that has been processed to remove noise and motion artifacts can be expressed as a time series: X(t) = {x1(t), x2(t),...x j (t),…,x n (t)}, where t represents time, which can be N different moments, x j (t) represents the time series data of the HbO2 concentration change signal of the jth channel, j = 1, 2, ..., n.

[0051] Step S130 , performing feature extraction on the processed tactile force change signal to obtain tactile force feature parameters.

[0052] In one embodiment of the present invention, the tactile force characteristic parameters include maximum gripping force, force duration, force change rate and force stability.

[0053] Specifically, in this embodiment, the maximum gripping force refers to the maximum value of the tactile force signal after filtering within a period of time t, which is expressed by the formula F max = max(F(t)), where F(t) is the tactile force change signal within a period of time t. The duration of force T d It refers to the total time that the tactile force change signal exceeds the preset force value, which is expressed as T d =∑t∈T Δt, if F(t)≥F th , where T represents the total time of a single acquisition, Δt represents the time interval for acquiring the tactile force change signal, and F th Represents the preset force value. The rate of change of force can be estimated by the time derivative of the tactile force change signal, using the differential approximation of discrete time points, expressed as The stability of the force can be evaluated by the standard deviation of the tactile force change signal within a given period of time, which is expressed as in, is the average amount of tactile force change signal, and the formula is F i It represents the value of the tactile force signal at the i-th sampling point at a certain moment, i=1, 2, ..., N, and N represents the total number of sampling points, that is, N different moments.

[0054] Step S140, selecting some channels with higher attention importance scores in the processed brain blood oxygen change signal through a self-attention mechanism.

[0055] In step S110 of the embodiment of the present invention, a multi-channel cerebral blood oxygen change signal is acquired by a multi-channel near-infrared brain function imaging device. Even after the processing of step S120, the multi-channel signal still has the characteristics of large data volume and complex calculation. Some existing analysis methods usually perform calculations and analyses based on the data of all channels, which may not only contain redundant information, but also greatly increase the calculation burden. Therefore, in order to improve the analysis efficiency, the method of the embodiment of the present invention introduces a graph convolutional network (GAT) with a self-attention mechanism to analyze the multi-channel cerebral blood oxygen change signal, so as to dynamically identify and select channels with high information content, that is, to complete channel selection, so that the subsequent steps can perform further feature extraction and evaluation and analysis of the brain function of the subject based on the selected high-information channels.

[0056] In one embodiment of the present invention, the method of selecting some channels with higher attention importance scores in the processed cerebral blood oxygen change signal by the self-attention mechanism specifically includes the following steps:

[0057] Step S141, based on the Pearson correlation coefficient, the correlation between the processed cerebral blood oxygen change signals of every two channels in the multi-channel is calculated, and based on the correlation, an adjacency matrix corresponding to all channels in the cerebral blood oxygen change signal is constructed, wherein if the correlation is greater than the Pearson correlation coefficient threshold, the element value at the corresponding channel position in the adjacency matrix is ​​1, otherwise the element value at the corresponding channel position in the adjacency matrix is ​​0, and the element values ​​on the main diagonal of the adjacency matrix are all 0.

[0058] Specifically, assuming that the time series data of the processed cerebral blood oxygen change signals of two channels are x and y, respectively, the Pearson correlation coefficient between the cerebral blood oxygen change signals of the two channels is calculated by the following formula: Among them, x t and t are the values ​​of the time series data x and y of the cerebral blood oxygen change signal at time t, and are the average values ​​of the time series data x and y, respectively. The larger the Pearson correlation coefficient, the higher the synchronization degree and the greater the correlation of the data of the two channels. This can determine the linear dependence between all channels. Then, since the Pearson correlation coefficient is symmetrical, that is, ρ xv =ρ vx , so the adjacency matrix corresponding to all channels can be constructed by combining the following formula: For example, when a 63-channel near-infrared brain function imaging device is used to collect 63-channel brain blood oxygen change signals, the adjacency matrix constructed here is a 63×63 adjacency matrix. In the above formula, the correlation between the two channels above the Pearson correlation coefficient threshold is significant, that is, the element value at the corresponding position in the adjacency matrix is ​​marked as 1, and the correlation between the two channels below or equal to the Pearson correlation coefficient threshold is not significant, that is, the element value at the corresponding position in the adjacency matrix is ​​marked as 0, and in order to eliminate autocorrelation, the element values ​​on the main diagonal of the adjacency matrix are all 0, thereby constructing the adjacency matrix. Among them, the larger the Pearson correlation coefficient threshold, the fewer the number of correlated connections between channels with significant correlation, and vice versa, the more the number of uncorrelated connections between channels with insignificant correlation.

[0059] Step S142, constructing a graph based on the adjacency matrix, the graph including nodes and edges for representing the connection relationship between the nodes, the processed cerebral blood oxygen change signals of the corresponding channels with an element value of 1 in the adjacency matrix serving as the nodes of the graph, and the correlations between the processed cerebral blood oxygen change signals of the corresponding channels serving as the edges of the graph.

[0060] Specifically, the adjacency matrix constructed in step S141 is used to construct a graph G, G = (V, E), wherein V represents a set of nodes in the graph, each node represents the time series data of the cerebral blood oxygen change signal of each channel corresponding to an element value of 1 in the adjacency matrix, and E represents a set of edges in the graph, each edge represents the correlation connection between the time series data of the cerebral blood oxygen change signal of each channel corresponding to an element value of 1 in the adjacency matrix, which is specifically represented by the correlation between the two channels.

[0061] Step S143, input the graph into the graph convolutional network of the self-attention mechanism, so that the graph convolutional network of the self-attention mechanism outputs multiple attention weights corresponding to each node, and obtains the attention importance score of each node in the graph based on the multiple attention weights corresponding to each node, and the attention weights are used to represent the importance of each node to each of its neighbor nodes.

[0062] Specifically, this embodiment introduces a self-attention mechanism, that is, uses a graph convolutional network with a self-attention mechanism to learn the graph, and can automatically assign different weights to each node in the graph to represent the importance of the node to each of its neighboring nodes, so that channels with high information content can be dynamically screened out.

[0063] In GAT, the self-attention layer calculates the attention weight of each node on the graph to its neighbor nodes using the following formula: α xz =softmax(LeakyReLU(a T [Wh x ||Wh z ])), where α xz represents the attention weight of node x to its neighbor node z, h x represents the feature vector of node x, that is, the feature obtained by extracting the cerebral blood oxygen change signal x of the corresponding channel, W represents the shared linear transformation matrix for feature learning, and the feature vector h x The converted feature is represented as Wh x ,h z The feature vector representing the neighbor node z of node x includes the features obtained by extracting the cerebral blood oxygen change signal z of the corresponding channel. The feature vector h z The converted feature is represented as Wh z , a represents the attention mechanism parameter, LeakyReLU is a nonlinear activation function used to avoid overfitting, and the softmax normalization function ensures that the sum of the attention weights from node x to all its neighboring nodes z is 1, making the influence and attention distribution of the graph convolutional network model of the self-attention mechanism between nodes more reasonable. Then, the attention importance score of each node on the graph is output through the following formula: S x =||h′ x ||2, where h′ x =∑ z α xz h z , from this formula, we can see that through the attention weight α xz Aggregate the features of all neighboring nodes z of node x to update the feature representation of node x. The purpose is to obtain a new feature representation h′ of node x by weighting the features of neighboring nodes. x, so that it contains the information of neighbor nodes. x It reflects the overall importance of node x in the entire graph (i.e., the attention or focus that node x obtains from all its neighboring nodes z on the graph). A higher score indicates that the corresponding node x (the time series data of the cerebral blood oxygen change signal of the corresponding channel) has a higher amount of information.

[0064] Step S144, sorting the attention importance scores of the nodes in order of high and low, and selecting some channels with higher attention importance scores.

[0065] Specifically, the attention importance scores of each node on the graph obtained in step S143 are sorted in order from high to low or from low to high, and then a portion of channels corresponding to higher-ranked attention importance scores can be selected, for example, the channels ranked in the top 70% of the attention importance scores. This is only an example, and different numbers of channels with higher scores can be specifically selected according to the set threshold.

[0066] Step S150, performing feature extraction on the cerebral blood oxygen change signals of some channels with higher attention importance scores to obtain cerebral blood oxygen feature parameters.

[0067] In one embodiment of the present invention, the cerebral blood oxygen characteristic parameters include the cerebral blood oxygen activation level and the clustering coefficient and global efficiency of the brain functional network connection level.

[0068] Specifically, since the cerebral blood oxygen change signal collected in step S110 includes a task-state cerebral blood oxygen change signal and a resting-state cerebral blood oxygen change signal, the cerebral blood oxygen activation level, i.e., the brain activation assessment, is obtained by calculating the difference between the task-state cerebral blood oxygen signal and the resting-state cerebral blood oxygen signal, specifically, the activation level A is calculated by the cerebral blood oxygen data of each channel in the selected high-information channel set (partial channel) S. j (t) The change relative to the baseline can be expressed as: in is the baseline blood oxygen level of channel j, that is, the average value during the resting state, X j (t) is the task-state blood oxygen level of channel j. For the brain functional network connectivity level, i.e., brain functional network connectivity assessment, based on the selected high-information channel set (partial channels), the phase locking value (PLV) used to analyze the phase synchronization between brain blood oxygen change signals is selected to construct the brain functional connection network, and the brain functional network connectivity level is evaluated by calculating the clustering coefficient and global efficiency. For each channel signal x j (t) Apply Hilbert transform to extract instantaneous phase φ j(t) and then calculate the PLV between the brain blood oxygen signals of each channel. For example, for channels j and k, the PLV formula is: Construct the adjacency matrix B based on the calculated PLV values: The channels with element values ​​of 1 in the adjacency matrix are taken as nodes, and the channels with element values ​​of 1 are connected to form edges. The channels with element values ​​of 0 are not connected to form edges, thereby obtaining a brain function connection network. The clustering coefficient calculation formula is: The global efficiency calculation formula is: Among them, channels j and k are nodes j and k, T j is the number of neighbor edges of node j, K j is the degree of node j, M is the total number of nodes in the network, and d(j,k) is the shortest path length between nodes j and k.

[0069] Step S160, obtaining a finger force evaluation result based on the tactile force characteristic parameter, and obtaining a brain function evaluation result based on the cerebral blood oxygen characteristic parameter, thereby realizing a rehabilitation evaluation of the fine motor function of the hand.

[0070] In one embodiment of the present invention, obtaining the finger force evaluation result based on the tactile force characteristic parameter in step S160 specifically includes the following steps:

[0071] By comparing and analyzing the tactile force characteristic parameters with the corresponding health indicators of healthy people of the same gender and age group as the test subject, and comparing and analyzing the tactile force characteristic parameters obtained at different rehabilitation stages of the test subject, the finger force evaluation result of the test subject is obtained; the finger force evaluation result includes the grade assessment of finger force control ability and finger force sensitivity, the finger force control ability includes the maximum gripping force and the duration of the force, and the finger force sensitivity includes the rate of change of the force and the stability of the force.

[0072] Specifically, the maximum grip force F of the subject is max , the duration of force T d , the rate of change of force R(t) and the stability of force σ FThe patients were compared with the corresponding health indicators of healthy people of the same age and gender, and were rated into four levels according to the corresponding ratios: D, C, B, and A. Among them, D means that the rehabilitation performance of the tested person is lower than 25% of the average level of healthy people, C means that the rehabilitation performance of the tested person is between 25% and 50% of the average level of healthy people, B means that the rehabilitation performance of the tested person is between 50% and 75% of the average level of healthy people, and A means that the rehabilitation performance of the tested person reaches 75% and above of the average level of healthy people. Through quantitative evaluation, the relative performance of the patient in terms of strength control stability and sensitivity when performing fine hand movements can be clearly identified, which is conducive to targeted adjustment of rehabilitation strategies and ensuring that the rehabilitation plan can be effectively optimized according to the patient's rehabilitation situation. At the same time, through comparative analysis of parameter indicators obtained at different stages of patient rehabilitation, the progress of the patient's rehabilitation assessment can be timely understood, which is further conducive to adjusting rehabilitation plans and strategies.

[0073] In one embodiment of the present invention, obtaining the brain function assessment result based on the cerebral blood oxygen characteristic parameter in step S160 specifically includes the following steps:

[0074] By comparing and analyzing the clustering coefficient and global efficiency of the cerebral blood oxygen activation level and the brain functional network connectivity level with the corresponding health indicators of healthy people of the same gender and age group as the test subject, and comparing and analyzing the clustering coefficient and global efficiency of the cerebral blood oxygen activation level and the brain functional network connectivity level obtained at different rehabilitation stages of the test subject, a grade assessment of the cerebral blood oxygen activation level and the brain functional network connectivity level of the test subject is obtained.

[0075] Specifically, the level of brain functional network connectivity is evaluated by the clustering coefficient and global efficiency. The larger the values ​​calculated by the two indicators, the higher the level of brain functional network connectivity, and the closer it is to the corresponding health indicators of healthy people of the same gender and age group, that is, the better the recovery of the test subjects and patients. The evaluation of brain blood oxygen activation level and brain functional network connectivity level is similar to the evaluation method of the aforementioned finger strength evaluation results, so it will not be described in detail here.

[0076] In one embodiment of the present invention, the method further comprises the following steps:

[0077] Step S170, establishing a rehabilitation assessment database for the subject, saving the tactile force characteristic parameters, the cerebral blood oxygen characteristic parameters, the finger force assessment results and the brain function assessment results in real time in the rehabilitation assessment database, and displaying the rehabilitation assessment results of the subject's hand fine motor function in a report.

[0078] Specifically, if Figure 3As shown, the rehabilitation assessment parameters and results of the hand fine motor function of the test subject are displayed in the form of a report by using statistical methods such as line graphs and bar graphs, which intuitively provides feedback on the rehabilitation assessment of the hand fine motor function. The database stores the finger force assessment results and brain function assessment results of the test subject and the patient at different stages, as well as the assessment result data of different test subjects. It can also support data query, call and update, making it easier to understand the patient's rehabilitation status and progress at different stages of the entire rehabilitation cycle, as well as the patient's historical rehabilitation assessment results.

[0079] In one embodiment of the present invention, the method further includes the following contents:

[0080] While collecting data, the processed multi-channel tactile force change signals are transmitted using the TCP / IP protocol and the signal data is encoded into a byte stream, which is transmitted in real time through a WebSocket connection to the human-computer interaction system interface developed by C#.NET and Unity3D for visual display. These byte streams are decoded and converted into visualization parameters. The tactile force data is linked to the hand model in the development platform model library, which can intuitively and dynamically display the hand movement status of the test subject when performing fine hand movements, and provide real-time feedback on the changes in the patient's finger tactile force. The tactile force data determines the degree of curvature of the displayed hand, etc., which produces visual stimulation to the patient to ensure the patient's good condition and drive hand movement, which can further improve the accuracy of the evaluation.

[0081] Corresponding to the above method, the present invention also provides a hand fine motor function rehabilitation assessment system, which includes a computer device, the computer device includes a processor and a memory, the memory stores computer instructions, the processor is used to execute the computer instructions stored in the memory, and when the computer instructions are executed by the processor, the system implements the steps of the above method.

[0082] The embodiment of the present invention also provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the steps of the aforementioned method are implemented. The computer-readable storage medium can be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the technical field.

[0083] An embodiment of the present invention further provides a computer program product, comprising computer instructions, which implement the steps of the aforementioned method when executed by a processor.

[0084] It should be understood by those skilled in the art that the exemplary components, systems and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software or a combination of the two. Whether it is performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link via a data signal carried in a carrier.

[0085] It should be clear that the present invention is not limited to the specific configuration and processing described above and shown in the figures. For the sake of simplicity, a detailed description of the known method is omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown, and those skilled in the art can make various changes, modifications and additions, or change the order between the steps after understanding the spirit of the present invention.

[0086] In the present invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with features of other embodiments or replace features of other embodiments.

[0087] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the embodiments of the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for evaluating the rehabilitation of hand fine motor function, characterized in that: The method comprises: Collect the tactile force change signal and multi-channel brain blood oxygen change signal of the test subject when performing hand test actions; Performing adaptive sliding average filtering on the tactile force change signal, and removing noise and motion artifacts from the multi-channel cerebral blood oxygen change signal; Extracting features of the processed tactile force change signal to obtain tactile force feature parameters; The self-attention mechanism is used to select some channels with higher attention importance scores in the processed brain blood oxygen change signal; Feature extraction is performed on the cerebral blood oxygen change signals of some channels with higher attention importance scores to obtain cerebral blood oxygen feature parameters; A finger force evaluation result is obtained based on the tactile force characteristic parameter, and a brain function evaluation result is obtained based on the cerebral blood oxygen characteristic parameter, thereby realizing a rehabilitation evaluation of the fine motor function of the hand.

2. The method according to claim 1, characterized in that: The self-attention mechanism is used to select some channels with higher attention importance scores in the processed brain blood oxygen change signal, including: Calculate the correlation between the processed cerebral blood oxygen change signals of every two channels in the multi-channels based on the Pearson correlation coefficient, and construct an adjacency matrix corresponding to all channels in the cerebral blood oxygen change signal based on the correlation, wherein if the correlation is greater than the Pearson correlation coefficient threshold, the element value at the corresponding channel position in the adjacency matrix is ​​1, otherwise the element value at the corresponding channel position in the adjacency matrix is ​​0, and the element values ​​on the main diagonal of the adjacency matrix are all 0; Constructing a graph based on the adjacency matrix, the graph including nodes and edges for indicating connection relationships between nodes, wherein the processed cerebral blood oxygen change signals of the corresponding channels whose element values ​​in the adjacency matrix are 1 are used as nodes of the graph, and the correlations between the processed cerebral blood oxygen change signals of the corresponding channels are used as edges of the graph; Inputting the graph into a graph convolutional network of a self-attention mechanism, so that the graph convolutional network of the self-attention mechanism outputs a plurality of attention weights corresponding to each node, and obtaining an attention importance score of each node in the graph based on the plurality of attention weights corresponding to each node, wherein the attention weight is used to indicate the importance of each node to each of its neighboring nodes; Sort the attention importance scores of each node in order of high to low, and select some channels with higher attention importance scores.

3. The method according to claim 1, characterized in that The tactile force characteristic parameters include the maximum grip force, force duration, force change rate and force stability; the cerebral blood oxygen characteristic parameters include the cerebral blood oxygen activation level and the clustering coefficient and global efficiency of the brain functional network connection level.

4. The method according to claim 3, characterized in that The obtaining of the finger force evaluation result based on the tactile force characteristic parameter comprises: By comparing and analyzing the tactile force characteristic parameters with the corresponding health indicators of healthy people of the same gender and age group as the testee, and comparing and analyzing the tactile force characteristic parameters obtained at different rehabilitation stages of the testee, the finger force evaluation result of the testee is obtained; the finger force evaluation result includes the grade assessment of the finger force control ability and the finger force sensitivity, the finger force control ability includes the maximum gripping force and the duration of the force, and the finger force sensitivity includes the rate of change of the force and the stability of the force; The obtaining of the brain function assessment result based on the cerebral blood oxygen characteristic parameter includes: By comparing and analyzing the clustering coefficient and global efficiency of the cerebral blood oxygen activation level and the brain functional network connectivity level with the corresponding health indicators of healthy people of the same gender and age group as the test subject, and comparing and analyzing the clustering coefficient and global efficiency of the cerebral blood oxygen activation level and the brain functional network connectivity level obtained at different rehabilitation stages of the test subject, a grade assessment of the cerebral blood oxygen activation level and the brain functional network connectivity level of the test subject is obtained.

5. The method according to claim 1, characterized in that The hand test movements include some or all of the following: thumb-to-finger, fingertip-to-finger, three-finger pinch, cylinder grip, side pinch and fist.

6. The method according to claim 1, characterized in that The tactile force change signal is obtained by collecting the tactile force change signals of the thumb, index finger, middle finger, ring finger and little finger through a capacitive pressure sensor, and the multi-channel cerebral blood oxygen change signal is obtained by collecting through a multi-channel near-infrared brain functional imaging device covering the prefrontal lobe, premotor area, supplementary motor area, primary motor area, sensorimotor area and occipital lobe.

7. The method according to claim 1, characterized in that The method further comprises: A rehabilitation assessment database for the test subject is established, the tactile force characteristic parameters, the cerebral blood oxygen characteristic parameters, the finger force assessment results and the brain function assessment results are stored in the rehabilitation assessment database in real time, and the rehabilitation assessment results of the test subject's hand fine motor function are displayed in a report.

8. A hand fine motor function rehabilitation assessment system, comprising a processor, a memory and computer instructions stored in the memory, characterized in that: The processor is used to execute the computer instructions, and when the computer instructions are executed, the system implements the steps of the method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer program product comprising computer instructions, characterized in that When the computer instructions are executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Citation Information

Patent Citations

  • Near-infrared-based multi-band physiological signal feedback system and using method thereof

    CN111281399A

  • Evaluation method and system of near-infrared brain function and touch force / motion information fusion

    CN111631731A

  • Equipment control method based on motor imagery EEG (Electroencephalograph), and terminal

    CN113143295A

  • Multi-passband graph convolution fusion brain connection classification method

    CN114287908A

  • Dynamic rehabilitation evaluation method and system based on fNIRS brain function map

    CN115349857A