CN-122006116-A - Bone fracture rehabilitation treatment regulation and control system based on traditional Chinese medicine meridian electrical stimulation

CN122006116ACN 122006116 ACN122006116 ACN 122006116ACN-122006116-A

Abstract

The invention relates to the technical field of electrotherapy, in particular to a bone fracture rehabilitation treatment regulation system based on traditional Chinese medicine meridian electrical stimulation. The system comprises a data preparation module for obtaining multi-mode expression descriptors of a patient, wherein the multi-mode expression descriptors comprise physiological, behavioral and facial expression vectors, a reaction weight analysis module for clustering historical patients based on disease consistency to obtain a bone fracture approximate set, extracting a time sequence reaction descriptor and calculating reaction reference weights, and a stimulation control module for training a feedback prediction model based on bone fracture approximate set data and the reaction reference weights and determining acupoint combinations and current data for the patient to be stimulated by using the model. According to the invention, through multi-mode data fusion and time sequence feature extraction, the dynamic stress response of a patient is comprehensively quantified, and the parameter mismatch problem is reduced by combining with a personalized correction mechanism, so that the reliability and the safety of an electrical stimulation scheme are improved.

Inventors

SHEN ZHIZHONG
LI QI
ZHAI SHICONG
YU LIANG
Kou Hailong

Assignees

河南医药大学第一附属医院(河南省结核病医院)

Dates

Publication Date: 20260512
Application Date: 20260413

Claims (10)

1. A bone fracture rehabilitation therapy regulation and control system based on traditional Chinese medicine meridian electrical stimulation, which is characterized by comprising: the data preparation module is used for acquiring a performance descriptor of the patient at the moment under the stimulation of the current data of the acupoint combination from historical patient sample data, wherein the performance descriptor comprises a physiological performance vector, a performance vector and a facial performance vector; The reaction weight analysis module is used for clustering according to the condition consistency condition of cases among patients to obtain a bone fracture approximate set, extracting and combining time sequence characteristics according to the approximate condition of a performance descriptor of each patient at time to obtain a reaction descriptor of each patient at time sequence; the simulation control module is used for training to obtain a feedback prediction model based on the case data of each patient in the approximate set of bone injuries in the historical patient sample data and the current data and the simulation feedback data of the acupoint combination by taking the reaction reference weight as the weight, and determining the scheme of the acupoint combination and the current data by using the feedback prediction model based on the case data of the patient to be stimulated and the expected simulation feedback data.
2. The system for regulating and controlling rehabilitation of bone injury based on traditional Chinese medicine meridian electrical stimulation according to claim 1, wherein the method for acquiring the approximate set of bone injury comprises the following steps: extracting structural feature vectors of the pathological states of the bone fracture from the electronic medical record data of each patient, taking the similarity of the structural feature vectors between the two patients as a disease consistency index, clustering the patients by adopting a clustering algorithm based on the disease consistency index as a measurement to obtain a cluster as an approximate set of the bone fracture.
3. The system for regulating and controlling rehabilitation of bone fracture based on traditional Chinese medicine meridian electrical stimulation according to claim 1, wherein the method for acquiring response descriptors comprises the following steps: In the single complete stimulation of each patient, combining the similarity of the two time expression descriptors and the time difference between the time to obtain the characteristic approximation degree between the time, wherein the magnitude of the time difference is in negative correlation with the characteristic approximation degree; selecting a representation descriptor corresponding to the central moment of each time sequence approximate set as a characteristic descriptor; all feature descriptors are spliced in time sequence to generate a reaction descriptor.
4. The system for regulating and controlling rehabilitation of bone fracture based on electrical stimulation of meridian of traditional Chinese medicine according to claim 1, wherein the method for obtaining the response reference weight comprises the following steps: In the bone fracture approximate set, for any one patient, taking each other patient as an analysis patient in turn, and determining a stimulation condition consistency index according to the similarity between the patient and the analysis patient in the acupoint combination and the current data; analyzing the similarity degree of the response descriptors between the patient and the analysis patient, and combining the indexes of consistent stimulation conditions to obtain the correlation coefficient between the patient and the analysis patient; And combining the correlation coefficients of the patient and all other patients to obtain the response reference weight of the patient.
5. The system for regulating and controlling rehabilitation of bone fracture based on traditional Chinese medicine meridian electrical stimulation according to claim 4, wherein the method for acquiring the stimulation condition consistency index comprises the following steps: counting the number of the same acupoints in the acupoint combination between the patient and the analysis patient, and taking the number as the acupoint approximation degree; Combining the intensity difference and the frequency difference of the current data between the patient and the analysis patient to obtain a current approximation degree, wherein the intensity difference and the frequency difference are inversely related to the current approximation degree; and fusing the point approximation degree and the current approximation degree to obtain the index of consistent stimulation conditions between the patient and the analysis patient.
6. The system for regulating and controlling rehabilitation of bone fracture based on electrical stimulation of meridian of traditional Chinese medicine according to claim 4, wherein the method for obtaining the correlation coefficient comprises the following steps: And fusing the response related index and the stimulus condition consistent index between the patient and the analysis patient to obtain the related coefficient of the patient and the analysis patient.
7. The system for regulating and controlling rehabilitation of bone fracture based on traditional Chinese medicine meridian electrical stimulation according to claim 1, wherein the method for acquiring the feedback prediction model comprises the following steps: taking the case data of the patient and the current data of the acupoint combination as input, taking the stimulus feedback data as output, taking the corresponding reaction reference weight of the patient as weight, and training the patient sample data in each bone injury approximate set to obtain a feedback prediction model of each bone injury approximate set.
8. The system according to claim 1, wherein the scheme for determining acupoint combinations and current data by using a feedback prediction model based on case data and expected stimulation feedback data of a patient to be stimulated comprises: Extracting structural feature vectors of the pathological states of the bone injuries from case data of the patient to be stimulated, matching the structural feature vectors with structural feature vectors corresponding to the central points of each bone injury approximate set, and screening a bone injury approximate set with the closest vectors as a reference set of the patient to be stimulated; According to the similarity degree of the patient to be stimulated and each historical patient in the reference set in case data, screening a preset number of most similar historical patients from the reference set to form a personalized reference subset; For the current data of each candidate acupoint combination, screening the reference stimulation feedback data of patients with similar candidate acupoint combinations and current data in the personalized reference subset, and carrying out weighting correction on the basic prediction feedback data based on the reference stimulation feedback data to obtain target prediction stimulation feedback data of the current data of each candidate acupoint combination; and screening out acupoint combinations and current data of the patient to be stimulated as a final scheme according to the approximate conditions of the target predicted stimulation feedback data and the expected stimulation feedback data.
9. The system for regulating and controlling healing of bone injury based on electrical stimulation of meridian of traditional Chinese medicine according to claim 7, wherein the stimulation feedback data is time series statistics of sensory feeling of patients.
10. The system for regulating and controlling rehabilitation of bone fracture based on electrical stimulation of meridian of traditional Chinese medicine according to claim 1, wherein the method for acquiring the expression descriptor comprises the following steps: Synchronously acquiring multi-mode biological characteristic data of a historical patient in a stimulation process, wherein the multi-mode biological characteristic data at least comprises physiological signal data representing physical physiological states, behavioral action data representing limb stress responses and facial image data representing facial emotional states; And respectively carrying out feature coding and normalization processing on the physiological signal data, the behavior action data and the facial image data, and fusing coding features in the same time window to construct a performance descriptor at the moment.

Description

Bone fracture rehabilitation treatment regulation and control system based on traditional Chinese medicine meridian electrical stimulation Technical Field The invention relates to the technical field of electrotherapy, in particular to a bone fracture rehabilitation treatment regulation system based on traditional Chinese medicine meridian electrical stimulation. Background Neuromuscular electrical stimulation (NMES) and Transcutaneous Electrical Nerve Stimulation (TENS) are common aids in rehabilitation engineering by applying current pulses to specific muscle groups or nerve nodes (e.g., motor points corresponding to acupoints in chinese medicine) to induce muscle contraction or modulate nerve conduction. In practice, the device control parameters are usually set empirically by the operator, however, the bio-impedance and neural tolerance of the human body to electrical stimulation (Tolerance) are not constant, but are dynamically influenced by individual physical differences, pathological conditions (such as bone fracture level) and stimulation duration (tolerance drift). Because different patients respond to current stimulation and pain differently, current intensity and frequency are required to be adjusted according to the tolerance degree of the patients when the meridian is electrically stimulated (EA), deep nerves cannot be activated due to too small current, severe pain or muscle spasm is caused due to too large current, and compliance and safety of rehabilitation training are affected. The common treatment only depends on single current intensity feedback, and the parameter mismatch problem caused by physique difference and tolerance drift of patients in different treatment stages is difficult to quantify, so that the current condition of the patients when the meridians are electrically stimulated is difficult to accurately regulate and control. Disclosure of Invention In order to solve the technical problem that in the prior art, single current intensity feedback is difficult to quantify parameter mismatch caused by physique difference and tolerance drift of a patient in different treatment stages, the invention aims to provide a bone fracture rehabilitation treatment regulation and control system based on traditional Chinese medicine meridian electrical stimulation, and the adopted technical scheme is as follows: the invention provides a bone fracture rehabilitation therapy regulation and control system based on traditional Chinese medicine meridian electrical stimulation, which comprises the following components: the data preparation module is used for acquiring a performance descriptor of the patient at the moment under the stimulation of the current data of the acupoint combination from historical patient sample data, wherein the performance descriptor comprises a physiological performance vector, a performance vector and a facial performance vector; The reaction weight analysis module is used for clustering according to the condition consistency condition of cases among patients to obtain a bone fracture approximate set, extracting and combining time sequence characteristics according to the approximate condition of a performance descriptor of each patient at time to obtain a reaction descriptor of each patient at time sequence; the simulation control module is used for training to obtain a feedback prediction model based on the case data of each patient in the approximate set of bone injuries in the historical patient sample data and the current data and the simulation feedback data of the acupoint combination by taking the reaction reference weight as the weight, and determining the scheme of the acupoint combination and the current data by using the feedback prediction model based on the case data of the patient to be stimulated and the expected simulation feedback data. Further, the method for obtaining the bone fracture approximate set comprises the following steps: extracting structural feature vectors of the pathological states of the bone fracture from the electronic medical record data of each patient, taking the similarity of the structural feature vectors between the two patients as a disease consistency index, clustering the patients by adopting a clustering algorithm based on the disease consistency index as a measurement to obtain a cluster as an approximate set of the bone fracture. Further, the method for obtaining the reaction descriptor comprises the following steps: In the single complete stimulation of each patient, combining the similarity of the two time expression descriptors and the time difference between the time to obtain the characteristic approximation degree between the time, wherein the magnitude of the time difference is in negative correlation with the characteristic approximation degree; selecting a representation descriptor corresponding to the central moment of each time sequence approximate set as a characteristic descriptor; all feature descriptors are