TW201445493A - A self-care system for assisting quantitative assessment of rehabilitation movement - Google Patents

Abstract

The present invention is related to a self-care system for assisting quantitative assessment of rehabilitation movement, comprising: (1) motion signals acquisition module; (2) movement parameters computing module; (3) templates construction module; (4) template matching module; and (5) rehabilitation progress monitoring module. Through the use of the system, patient will effectively improve the home self-monitoring and control extraordinary warning of self-care. Therefore, the invention will provide cloud industry, medical equipment development, health care providers or service providers, and research in the fields of care aids etc., with the required equipment and health-care service delivery systems.

Description

Autonomous rehabilitation motion quantitative evaluation system

The invention relates to a quantitative evaluation system for autonomous rehabilitation actions, which can effectively improve the self-monitoring and self-care warning benefits of the home end users.

With the advancement of domestic medical technology and the improvement of living standards, the average life expectancy of Chinese people has prolonged, causing an increase in the number of elderly people over the age of 65 in China. Although longevity is a long-cherished wish of human beings, the inevitable deterioration of limb function and chronic diseases accompanying the increase in age make the quality of life of the elderly seriously threatened.

Degenerative arthritis is the most common joint disease in the world, and its high prevalence rate has led to a significant increase in social resources and economic expenditure, not only in the increase of medical resources, medicines, surgery costs, but also because of work disability. The reduction of resources indirectly causes an overall loss. More than half of the population over 65 years of age develop degenerative arthritis, and the prevalence increases with age. In addition, the fifty shoulders have always been an important problem for middle-aged people. Among them, the recurrent 50 shoulders are mainly caused by trauma to the shoulders; the primary ones are caused by unexplained causes of joint deterioration. Therefore, it is quite difficult to prevent . Fifty shoulder patients often suffer from shoulder pain, which causes many functional movements to be restricted, which further leads to worsening of the condition.

The goal of rehabilitation medicine is to habituate, restore, and compensate for the missing limb function of the patient, so that it can return to society with confidence and live an independent life like a normal person. Rehabilitation exercise treatment is mostly performed by a rehabilitation physician or physical therapist. The hospital conducts free-handed guidance and combined rehabilitation equipment treatment, while the milder patients can follow the doctor's guiding principles to work hard at home. Clinically common exercise training methods, mainly through the assistance of mechanical equipment and professionals, perform active or passive traction exercises, such as leg isokinetic strength training instruments, and also have research and development related rehabilitation robots at home and abroad. The automatic control function drives the patient to perform non-autonomous muscle stretching exercises, so that patients who cannot exercise autonomously use rehabilitation equipment for rehabilitation. In addition to their complementary functions, these instruments provide kinematic or kinetic data for evaluation by physicians and therapists, but advanced equipment is not readily available for home use. In addition, limited physical therapists are difficult to meet a large number of patient needs, and it is more difficult to receive guidance or correction from a physical therapist anytime, anywhere. Therefore, the development of home-based simple rehabilitation equipment, combined with physical sensing devices to quantify the rehabilitation movement and provide immediate feedback guidance, will inevitably become a medical strategy worthy of development. In the future, the home care market is bound to develop rapidly, not only saving round trips. The cost of medical institutions and the heavy medical staff are of the highest value in rehabilitation applications.

Analysis of the Republic of China patents with high relevance to this case, most of the relevant inventions Research and development of rehabilitation equipment for mechanical and institutional linkages; some inventions combine virtual reality or games to simulate test scenarios, and then connect devices to generate feedback actions, but their main functions are still based on providing dynamic parameter measurements. Follow-up clinical evaluation.

The Republic of China Patent No. I377055 discloses an interactive limb movement The guiding method and the interactive limb motion rehabilitation system and the computer readable storage medium have the main technical features similar to the present invention, but the prior case does not disclose the algorithm of the physical sensor and processing parameters and the rehabilitation function. Sexual associations, etc. Even if combined with the Republic of China Patent No. 151919: a hand-held motion control diagnostic evaluation of rehabilitation gloves, only the information provided for the application of limbs or hand grasping is evaluated separately, and it is not easy to derive the main technical features of the case.

At present, the first intervention in the clinic is still the main intervention and associated with Physical therapy instruments have been developed. In recent years, the rapid development of micro-electromechanical and electromechanical integration technologies, robotic arms or rehabilitation machines began to introduce clinical practice, becoming another highly integrated and highly functional alternative rehabilitation instrument. Most of the master's research literatures are mainly in this field, mainly based on kinetic parameter measurement analysis and limb movement trajectory. The introduction of rehabilitation robot equipment is still in the early stage in Taiwan, and relevant research publications focus on the evaluation of the effectiveness of specific areas or ethnic groups. In recent years, due to the improvement of overall quality of life and the emphasis on medical quality, home care and management related issues Gradually develop to improve patient self-care and self-management effectiveness.

The system architecture, motion capture, sequence segmentation, comparison algorithm and its application at the home end are obviously different from the previous case. The user needs to establish and verify the guidance under the guidance of the medical staff. The waveforms of the individual rehabilitation demand action attributes are processed to quantify the action model; the user performs the similarity comparison of the time length, the joint displacement and the joint bending angle according to the action pattern database in the self-recovery stage; According to the standard deviation of the length of the action and the swing distance, the analysis of the speed and the high and low of the output action is given. The invention not only provides the evaluation of the functional quantitative indicators, but also displays the information of the correctness comparison and the guidance in the self-rehabilitation stage, thereby reducing the secondary injury caused by the incorrect positioning action at the home end. The main technical features proposed are not easily exemplified by those with ordinary knowledge in the technical field, and have the effect of improving self-rehabilitation.

The invention mainly provides a scientific and technical auxiliary nursing strategy for the prevention and treatment of joint diseases, among which the degenerative knee arthritis and the upper limbs are the most common.

In response to the above needs, the present invention proposes a quantitative evaluation system for autonomous rehabilitation actions. The system mainly includes: (1) action signal acquisition module, (2) motion parameterization calculation module, (3) action template establishment module, (4) action score comparison module, and (5) complex The health record record module, etc., analyzes the quantitative parameter values such as the length of the rehabilitation action, the displacement of the joint position and the change of the joint bending angle, and provides the establishment of the action template database; according to the standard deviation of the action time length and the swing distance, The analysis of the speed of the output action and the high and low analysis suggestion, the establishment of the action history record database, the two databases are combined to form a rehabilitation course database, which can be stored to portable memory devices, smart phones, tablet devices and cloud hard drives. Etc., to provide medical staff for follow-up rehabilitation tracking assessment. Through the invention, an independent rehabilitation action record and quantitative evaluation assistant system can be provided, so that the medical staff can grasp the immediate condition of the patient at any time, and give the patient the most appropriate adjustment and suggestion, thereby improving the treatment situation of the patient; In addition, it will effectively improve the self-monitoring and self-care warnings of home users to improve medical quality.

The present invention provides an autonomous rehabilitation action quantification evaluation system, comprising: (1) an action signal acquisition module for detecting at least one action change of a user by at least one sensor to obtain a repetition a signal sequence for performing the action; (2) an action parameterization calculation module connected to the action signal acquisition module for performing the signal sequence of the repeated measurement action obtained by the motion signal acquisition module The action sequence is segmentally cut into a plurality of sub-paragraphs and calculates the length of time of all the action sub-paragraphs and the normalized action change value; (3) an action template building module is connected to the action parameterized calculation module for The action sub-paragraph and its parameter values obtained under the guidance of the medical staff are stored in a rehabilitation treatment database; (4) an action score comparison module, connected to the motion parameterization calculation module and the action The template building module is configured to select the closest sub-action paragraph template from the rehabilitation treatment database according to the length of time in the self-rehabilitation stage, and perform action variation parameter distance scoring; and (5) a rehabilitation treatment Recording module, connected to the comparison operation scoring module for outputting operation The results are analyzed and returned to the rehabilitation treatment database to provide medical staff with follow-up rehabilitation tracking assessments.

According to the present invention, the autonomous rehabilitation motion quantitative evaluation system, wherein at least one motion change of the user refers to a change in a human limb or joint position, wherein the joint position changes as a joint point displacement or a joint bending angle change.

In the preferred embodiment, the at least one sensor is a physical sensor or an optical sensor, and wherein the physical sensor is a speedometer. , gyroscope, altimeter or mobile device including the above physical sensor, smart phone or portable device; the optical sensor is a webcam (Webcam or IPcam), Microsoft Kinect or ASUS WAVI Xtion.

According to the autonomous rehabilitation motion quantitative evaluation system of the present invention, the motion sequence segmentation cutting system normalizes the signal sequence of the repeated measurement action and obtains the average value of the sequence, and the segmentation falls on the average line and the lower wave type. The conversion action sequence becomes a sequence of action sub-paragraphs.

The self-recovery action quantitative evaluation system according to the present invention, wherein the rehabilitation treatment database includes an action template database and an action history record library for storing to a portable memory device, a smart phone, a tablet device Or a cloud drive.

According to the autonomous rehabilitation motion quantitative evaluation system of the present invention, the motion analysis refers to the analysis of the speed and the high and low of the output action according to the standard deviation of the action time length and the swing distance.

100‧‧‧Autonomous Rehabilitation Action Quantitative Evaluation System

101‧‧‧Action signal acquisition module

102‧‧‧Action Parametric Computing Module

103‧‧‧Action template building module

104‧‧‧Action Score Comparison Module

105‧‧‧Rehabilitation treatment record module

201‧‧‧Action signal acquisition

202‧‧‧Sensing value normalization

203‧‧‧ Is the single action completed?

204‧‧‧Action parameter calculation

205‧‧‧The duration of the action exceeds the threshold?

206‧‧‧Storage template

207‧‧‧Continue rehabilitation exercises?

208‧‧‧ action sample database

209‧‧‧Acoustic history

210‧‧‧Rehabilitation Treatment Database

211‧‧‧The length of access time is most similar to the model

212‧‧‧ Action sub-paragraph score

213‧‧‧Action analysis and history

401‧‧‧Users

402‧‧‧Android platform

403‧‧‧Mixed-up engine

404‧‧ ‧ Acceleration regulations

405‧‧‧ mobile device

406‧‧‧Wireless Transmission Module

407‧‧‧MySQL Database

408‧‧‧PHP engine

409‧‧‧Server

410‧‧‧HTTP Agreement

411‧‧‧Network Module

412‧‧‧Mixed-up engine

413‧‧‧ display

414‧‧‧ Input device

415‧‧‧Workstation

416‧‧‧Windows platform

417‧‧‧Physician/Rehabilitation

FIG. 1 is a structural diagram of an autonomous rehabilitation action quantitative evaluation system according to the present invention.

2 is a flow chart of the rehabilitation action analysis of the present invention.

FIG. 3 is a schematic diagram of the displacement and motion sequence division of the normalized pendulum motion joint according to the present invention.

FIG. 4 is a structural diagram of a seated straight leg rehabilitation system based on a smart phone according to the present invention.

Figure 5 is a sitting posture recuperation rehabilitation assisting system of the present invention.

Fig. 6 is a waveform diagram showing the knee joint movement angle of the present invention.

Figure 7 is a rehabilitation treatment management interface of the present invention.

The following examples are not intended to limit the invention, but are merely representative of several aspects and features of the invention.

FIG. 1 is a schematic diagram of an autonomous rehabilitation motion estimation system 100 according to the present invention, which mainly includes: (1) an action signal acquisition module 101 for detecting a use by at least one sensor At least one action change is performed to obtain a signal sequence of the repeated test action; (2) an action parameterization calculation module 102 is connected to the action signal capture module 101 for using the action signal capture module The signal sequence of the repeated measurement operation obtained by the 101 is segmented by the action sequence to form a plurality of sub-paragraphs and calculates the time length of all the action sub-paragraphs and the normalized action change value; (3) an action template creation module 103, Connected to the action parameterization calculation module 102, for storing the action sub-paragraph and its parameter values obtained under the guidance of the medical staff to a rehabilitation treatment database; (4) an action score comparison The module 104 is connected to the action parameterization calculation module 102 and the action template establishment module 103 for selecting the closest sub-action paragraph template from the rehabilitation treatment database according to the length of time in the autonomous rehabilitation phase. And proceed The action change parameter distance score is obtained; and (5) a rehabilitation treatment record module 105 is connected to the action score comparison module 104 for outputting the action analysis result and returning it to the rehabilitation course database. To provide medical staff for follow-up rehabilitation tracking assessment.

The present invention utilizes physical sensing placed on a limb or joint of a body. And an optical sensor in a non-contact form with the body to capture a limb movement or joint point change of the user performing the rehabilitation process, wherein the physical sensor can be an accelerometer, a gyroscope, an altimeter, or the like The physical sensor device, smart phone or portable device; and the optical sensor can be a webcam (Webcam or IPcam), Microsoft Kinect or ASUS WAVI Xtion.

Figure 2 is a flow chart of the rehabilitation action analysis. Capture 201 modules through motion signals Controlling and recording the above-mentioned one or more sensor signals, the obtained repeated measurement action signal sequence needs to be normalized and processed to obtain the average value of the sequence, and the action sequence is converted by dividing the falling wave on the average line to become the action sequence. The sequence of the action sub-paragraphs is further calculated by the action parameterization calculation module to calculate the time lengths of all the action sub-paragraphs, the joint position displacements, and the joint bending angle changes. Fig. 3 is a schematic diagram showing the displacement of the joint point and the action sequence of the pendulum motion after normalization, wherein the horizontal axis is time; the vertical axis is the displacement change after normalization, and the vertical axis value is 0.5 as the average line, by judging the lower wave above the average line The type is used to cut the front swing or the back swing motion, as indicated by the symbol X on the figure, and further determine the peak value, as indicated by the symbol O on the figure.

In the reference model training phase, the user gets the action under the guidance of the medical staff. The sub-paragraph and its parameter values are stored in the action template database through the action template creation module; in the user self-rehabilitation phase, the action score comparison module is based on the time length from the action template database. In 208, the closest sub-action paragraph template is selected, and the Euclidean geometric distance calculation of the joint position displacement and the peak value of the joint bending angle is calculated, thereby providing a measurement basis for the motion analysis. For example, when the length of the swing time is less than the difference between the selected reference templates and falls within the ±1σ standard range (68%) of the sample database statistics, it is considered as a match; if "not", then the action Execution is too fast or too slow, reminding you to redo; then, judging the peak distance difference of the swing amplitude, If it is within the maximum difference of 20%, it is regarded as normal; between 20% and 50%, it is considered to be slightly larger or slightly smaller; if it is more than 50%, it is too large or too small, reminding that it needs to be redone. According to the standard deviation of the action time length and the swing distance, the analysis of the speed and the high and low of the output action is given, and the analysis result is returned to the action history record 209 through the rehabilitation process record module. The rehabilitation treatment database 210 composed of the above operation template database 208 and the operation history record 209 library can be stored in a portable memory device, a smart phone, a tablet device or a cloud hard disk, etc., to provide medical personnel for follow-up. Rehabilitation tracking assessment.

Based on the above-mentioned system architecture and method, with 50 shoulder rehabilitation and degenerative knee Two application examples, such as arthritis rehabilitation, specifically demonstrate the feasibility of the present invention. First of all, for the fifty shoulders, also known as frozen shoulder or adhesive joint capsule inflammation, in the early stage of the disease, the patient will feel severe pain in the shoulder, because the movement of the shoulder joint is less and less, resulting in more serious adhesion, which leads to more and more joint activity. small. After the pain begins to diminish, the patient needs to use rehabilitation to help the shoulders recover from the range of activities. However, when the patient is autonomously rehabilitated at home, the correct rehabilitation action cannot be performed correctly, so that the rehabilitation effect is not as expected and may cause secondary injury. Therefore, the present invention utilizes the Kinect device developed by Microsoft to analyze and dynamically generate two-dimensional coordinate data of the upper limb joint point through the image analysis, thereby using the Kinect device to quantify and analyze the self-recovery motion evaluation of the commonly used pendulum motion of the rehabilitation, so that the medical staff can grasp The patient's rehabilitation course and progress, the specific results are shown in Figure 3.

Another case is degenerative arthritis, the most common of the human joint diseases. It is directly related to ageing, and the cause of degenerative arthritis is the excessive wear and tear of articular cartilage. The main symptoms are pain, stiffness, swelling and deformation, causing patients to suffer a certain degree of life. . The invention utilizes a smart phone embedded with an accelerometer to measure the three-dimensional acceleration value, and obtains an angular tilt change through a trigonometric function operation, thereby quantifying the analytical complex The evaluation of the knee joint activity angle of the commonly used sitting posture straight leg lifting exercise. This embodiment is mainly implemented on the Android mobile phone platform. The system architecture is as shown in FIG. 4, and mainly includes a mobile device 405, a server 409, and a workstation 415.

Figure 5 shows the system prototype interface. The mobile device processor is held at a sampling frequency of 4 Hz. Continue to capture the acceleration data, and then convert the acceleration data into the knee joint motion angle value, as shown in Figure 6, and transmit the data to the server via the wireless network; the server is designed using the PHP programming language to further analyze each The highest point angle value of the secondary exercise and the average angular velocity value during the exercise, the information obtained is stored in the MySQL database; the data is transmitted back to the workstation through the network, and the home rehabilitation is generated through the Windows rehabilitation treatment management platform. The data and analysis results are presented to physicians and therapists, as shown in Figure 7.

When the home is rehabilitated, the mobile device automatically records and calculates the rehabilitation exercise during the process. Information, which can be used to know the rehearsal direct-holding angle and the rehabilitation average angular velocity information, and present them in a simple and easy-to-understand chart, in which the bar chart shows the direct-staying angle; the angle chart shows the angle information and the average The angular velocity trend can also be used to know the improvement of muscle strength. Rehabilitation physicians and therapists are provided as a diagnostic reference through the above analysis.

The above detailed description is specific to one of the possible embodiments of the present invention, The present invention is not intended to limit the scope of the invention, and the equivalents and modifications of the invention are intended to be included in the scope of the invention.

201‧‧‧Action signal acquisition

202‧‧‧Sensing value normalization

203‧‧‧ Is the single action completed?

204‧‧‧Action parameter calculation

205‧‧‧The duration of the action exceeds the threshold?

206‧‧‧Storage template

207‧‧‧Continue rehabilitation exercises?

208‧‧‧ action sample database

209‧‧‧Acoustic history

210‧‧‧Rehabilitation Treatment Database

211‧‧‧The length of access time is most similar to the model

212‧‧‧ Action sub-paragraph score

213‧‧‧Action analysis and history

Claims (9)

An autonomous rehabilitation action quantitative evaluation system includes: (1) an action signal acquisition module for detecting at least one motion change of a user by at least one sensor to obtain a repeated measurement action a sequence of signals; (2) an action parameterization calculation module connected to the action signal acquisition module for performing sequence sequence of the signal sequence of the repeated measurement action obtained by the motion signal acquisition module The segment is cut into a plurality of sub-paragraphs and the time length of all the action sub-paragraphs and the normalized action change value are calculated; (3) an action template building module is connected to the action parameterized calculation module for guiding by the medical staff The action sub-paragraph and its parameter values obtained are stored in a rehabilitation course database; (4) an action score comparison module is connected to the action parameterization calculation module and the action template is established. The group is configured to select the closest sub-action paragraph template from the rehabilitation course database according to the length of time in the self-rehabilitation stage, and perform action change parameter distance scoring; and (5) a rehabilitation treatment record module, Connected to the action than scoring module to output analysis results and actions to be saved back to the rehabilitation treatment database to provide medical follow-up assessment of the rehabilitation track. For example, the self-recovery action quantitative evaluation system of claim 1 is characterized in that at least one action change of the user refers to a change in the position of the human limb or joint. For example, the quantitative evaluation system for autonomous rehabilitation actions of claim 2, wherein the joint position changes as a change in joint displacement or joint bending angle. The autonomous rehabilitation action quantitative evaluation system of claim 1, wherein the at least one sensor is a physical sensor or an optical sensor. For example, the self-recovery action quantitative evaluation system of claim 4, wherein the physical sensor is an accelerometer, a gyroscope, an altimeter or a mobile device including the above physical sensor Set, smart phone or portable device. For example, the self-recovery action quantitative evaluation system of claim 4, wherein the optical sensor is a webcam (Webcam or IPcam), Microsoft Kinect or ASUS WAVI Xtion. The method for quantitatively evaluating an autonomous rehabilitation action according to claim 1, wherein the sequence of the action sequence is normalized by the signal sequence of the repeated measurement action, and the average value of the sequence is obtained, and the segmentation is performed by segmentation. The action sequence is converted to an action sub-paragraph sequence on the average line down-wave type. For example, the self-recovery action quantitative evaluation system described in claim 1, wherein the rehabilitation treatment database includes an action template database and an action history record library for storing to a portable memory device, and wisdom. Phone, tablet device or cloud drive. For example, the self-recovery action quantitative evaluation system described in claim 1 of the patent scope refers to the analysis of the speed and the high and low of the output action according to the standard deviation of the length of the action time and the swing distance.