Smart Rehabilitation Device

Abstract

The embodiments of the present application disclose a smart rehabilitation device, which aims to solve the problems of the prior art, such as the inability to adjust position and height, low evaluation accuracy, and low rehabilitation management efficiency. The device comprises a mobile carriage, an embedded host computer, a mobile unit, and a wearable unit, each attached to the mobile carriage. The mobile unit and the wearable unit are both connected to the embedded host computer, which is equipped with a rehabilitation training module, a rehabilitation evaluation module, a data collection module, a data processing module, a movement control module, and a storage module. The mobile unit includes rollers, an electric push rod, and a drive unit installed on the mobile carriage, and the drive unit is connected to the movement control module. The wearable unit includes rehabilitation training gloves and a sensory device, and both the rehabilitation training gloves and the sensory device perform rehabilitation training via the rehabilitation training module and transmit rehabilitation training data to the data collection module.

Description

[Technical Field]

[0001] The embodiments of the present application relate to the medical technical field, and more particularly to a smart rehabilitation device. [Background technology]

[0002] With the rise in people's living standards and the accelerating aging of the population, cardiovascular disease has become a major health threat, with stroke being a major cause of death and permanent disability. Clinical research and practice have proven that with rapid and effective rehabilitation training, approximately 90% of patients can regain some level of daily movement and independent living ability. Clinically, 0 to 6 months is considered the golden recovery period, but rehabilitation is a long-term process, and most patients should continue for more than a year. Furthermore, due to the large number of stroke patients and the severe shortage of rehabilitation medical resources, there is a great demand for smart management devices.

[0003] Current rehabilitation treatments generally involve one-on-one treatment with a rehabilitation therapist assisting the patient by hand or using mechanical assistive devices. This method has the following drawbacks: (1) low treatment efficiency and high labor costs; (2) rehabilitation evaluations are highly subjective, making it difficult to ensure the accuracy of rehabilitation training evaluation results; (3) the selection of training movements is not similar to daily life, making it difficult for patients to return to normal social life; and (4) the position and height of the equipment are fixed, making it difficult to use for patients who have difficulty moving around. Summary of the Invention [Problem to be solved by the invention]

[0004] The embodiments of the present application provide a smart rehabilitation device that can improve the efficiency of rehabilitation treatment for patients and the intelligence of the rehabilitation management device. [Means for solving the problem]

[0005] The technical solutions in the embodiments of the present application are as follows: A smart rehabilitation device, the device comprising: a mobile carriage; the mobile carriage includes an embedded host computer, a mobile unit, and a wearable unit, the mobile unit and the wearable unit are both communicatively connected to the embedded host computer; the embedded host computer includes a rehabilitation training module, a rehabilitation assessment module, a data collection module, a data processing module, a movement control module, and a storage module; The rehabilitation training module is for providing a scenario training game designed based on rehabilitation movements, and the scenario training game is for training the upper limbs, lower limbs and hands of a rehabilitation training patient; the rehabilitation evaluation module is for evaluating the motor functions of the upper limbs, lower limbs, and hands of the rehabilitation training patient; The data collection module is for collecting information about the rehabilitation training patient, the information including eigenvectors corresponding to multiple dimensions of the patient, each eigenvector representing patient information in each dimension, the dimensions including the patient's medical history file, lifestyle habits, environmental factors, and rehabilitation training status; the data processing module is used for: predicting the impact degree of each dimension of patient information on the patient's rehabilitation based on the eigenvector of each dimension, to obtain a first prediction vector corresponding to the eigenvector of each dimension, the first prediction vector being for indicating the impact degree of each dimension of patient information on the patient's rehabilitation, and feature combining each of the first prediction vectors to obtain a target vector; predicting the rehabilitation effect corresponding to each dimension of patient information based on the target vector, to obtain a second prediction vector, the second prediction vector being for indicating the rehabilitation effect corresponding to the patient information of each dimension; and generating prescription information for the patient's rehabilitation based on the second prediction vector, the prescription information including medication information and drug dosage; the movement control module acquires the position of the rehabilitation training patient and controls the moving carriage to move its position and adjust its height by measuring the distance between the rehabilitation training patient and the moving carriage; The moving unit includes a roller, an electric push rod, and a driving device installed on the moving carriage, the driving device being connected to the movement control module, and the movement control module driving the roller by the driving device to move and adjust the position, and controlling the electric push rod to adjust the height; The wearable unit includes a rehabilitation training glove and a sensory device, both of which perform rehabilitation training via the rehabilitation training module and transmit rehabilitation training data to the data collection module.

[0006] Further, the data collection module includes a receiver and a camera, and the receiver is wirelessly connected to the rehabilitation training glove and the haptic device, respectively; the receiver is for receiving motion data of the rehabilitation training gloves and the sensory device, and the camera captures the motion of the rehabilitation training patient to collect the motion data; The data processing module is further for processing the motion data by multi-source information fusion.

[0007] Furthermore, for the man-machine interactive support, a card reading area and a button area are provided on the man-machine interactive support, the card reading area is for identifying the rehabilitation training patient individually, and a plurality of buttons are provided on the button area, and the buttons are for assisting the progress of the scenario training game; Regarding the cloud server, the exercise prescription is stored in the cloud server, The embedded host computer is for downloading the exercise prescription or uploading the rehabilitation training data via the cloud server; Regarding the power supply module, the power supply module includes a magnetic core, a transformer, a filter and a data line, and is used for supplying power and processing power signals.

[0008] Furthermore, the rehabilitation training glove includes a glove body and a first housing attached to the glove body, a first battery, a first circuit board and a bending sensor installed in the first housing, and a first indicator lamp installed in the first housing.

[0009] The haptic device further includes a band and a second housing attached to the band, a second circuit board and a second battery installed in the second housing, and a second indicator lamp installed in the second housing.

[0010] Further, the rehabilitation evaluation module includes an evaluation auxiliary unit and an evaluation unit, and the evaluation unit includes a smart scale evaluation subunit and a compensatory movement quantification evaluation subunit; The evaluation assistance unit is for completing a rehabilitation training movement identification task and obtaining an identification result; The evaluation unit is for obtaining a corresponding evaluation result based on the identification result. Furthermore, the embedded host computer can be used for remote medical diagnosis or online rehabilitation training guidance, realizing interconnection of terminals via a network.

[0011] Furthermore, a radar detector and a positioning module are connected to the movement control module, the radar detector is for detecting the position of the rehabilitation training patient, and the positioning module is for determining the current position of the mobile carriage.

[0012] Furthermore, the mobile carriage includes a base and a connection block, the roller is installed at the bottom of the base, the electric push rod is installed in the connection block, and the embedded host computer is connected to the top of the electric push rod.

[0013] Furthermore, an infrared sensor and an infrared signal transmitter are connected to the movement control module, and the device realizes position and height adjustment by the infrared sensor, the infrared signal transmitter and the movement control module when in use. [Effects of the Invention]

[0014] Compared with the prior art, the advantageous effects of the embodiments of the present application are as follows:

[0015] First, an embodiment of the present application provides a smart rehabilitation device, which includes a mobile platform, an embedded host computer, a mobile unit, and a wearable unit, which are respectively attached to the mobile platform, and the embedded host computer controls the mobile unit to realize smart position / height identification and height adjustment functions; Second, an embodiment of the present application provides a smart rehabilitation device, which performs rehabilitation training tasks in the form of a virtual game through a rehabilitation training module in an embedded host computer, and obtains visualized quantified rehabilitation evaluation results through a rehabilitation evaluation module, and is easy to operate, has high operating efficiency, and is highly entertaining. Third, the embodiments of the present application provide a smart rehabilitation device, which can use rehabilitation training gloves and a sensory device to perform rehabilitation training for the upper limbs, lower limbs, and fingers in multiple directions. The device can identify the rehabilitation training movements and establish scale evaluation and compensation evaluation to form an accurate evaluation of the rehabilitation training movements. Compared with conventional evaluation systems and devices, the evaluation results are more accurate, and the evaluation model can be further trained, which has a wide application market. Fourth, the embodiment of the present application provides an intelligent rehabilitation device, which establishes a rehabilitation file through a man-machine interactive platform, and can grasp the rehabilitation progress of all patients at any time, improving the user experience. In addition, the detailed management file can reduce the burden on operators and make the entire rehabilitation training process more standardized. Fifth, an embodiment of the present application provides a smart rehabilitation device, which uses a rehabilitation training module to load various rehabilitation training games and select interesting games that are close to daily life to help patients with their rehabilitation training. The rehabilitation training module can fully train the patient's upper limbs, lower limbs, and fingers, accelerating the patient's rehabilitation progress. Sixth, an embodiment of the present application provides a smart rehabilitation device, which can obtain each dimensional information of a patient, predict the rehabilitation effect of the patient, and generate prescription information for the patient's rehabilitation based on each dimensional information of the patient, thereby improving the rehabilitation efficiency of the patient. [Brief explanation of the drawings]

[0016] [Figure 1] FIG. 1 is a first structural diagram of a smart rehabilitation device according to an embodiment of the present application. [Figure 2] FIG. 2 is a second structural diagram of the smart rehabilitation device according to the embodiment of the present application. [Figure 3] FIG. 3 is a third structural diagram of the smart rehabilitation device according to the embodiment of the present application. [Figure 4] FIG. 4 is a schematic diagram of a smart rehabilitation device according to an embodiment of the present application. [Figure 5] FIG. 5 is a structural diagram of gloves for rehabilitation training in a smart rehabilitation device according to an embodiment of the present application. [Figure 6] FIG. 6 is a structural diagram of a sensory device in a smart rehabilitation device according to an embodiment of the present application. [Figure 7] FIG. 7 is a fourth structural diagram of the smart rehabilitation device according to the embodiment of the present application. DETAILED DESCRIPTION OF THE INVENTION

[0017] In order to make the objectives, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings of the embodiments of the present application. Of course, the embodiments described below are only a part of the embodiments of the present application, but not all of the embodiments. In general, the components of the embodiments of the present application described and shown in the drawings herein may be arranged and designed in a variety of different configurations.

[0018] Therefore, the detailed description of the embodiments of the present application provided below with reference to the drawings is only intended to illustrate selected embodiments of the present application, and does not limit the scope of the claims of the present application. Based on the embodiments of the present application, any other embodiments that a person skilled in the art can obtain without inventive efforts fall within the scope of protection of the present application.

[0019] It should be understood that in describing the example embodiments of the present application, the terms "first," "second," etc. are for descriptive purposes only and should not be understood as indicating or implying the relative importance or number of technical features being indicated, such that a feature qualified by "first," "second," etc. may explicitly or implicitly include one or more of said features.

[0020] In describing the embodiments of the examples of the present application, it should be noted that unless otherwise clearly specified and limited, the terms "installation," "mounting," "connection," and "connection" should be understood in a broad sense, and may refer to, for example, a fixed connection, a detachable connection, or an integral connection, a mechanical connection, an electrical connection, or the ability to communicate with each other, a direct connection, an indirect connection through an intermediate element, internal communication between two elements, or an interactive relationship between two elements. Those skilled in the art can understand the specific meanings of the above terms in the embodiments of the examples of the present application according to specific circumstances.

[0021] FIG. 1 is a block diagram of the overall structure of a smart rehabilitation device according to an embodiment of the present application. Referring to FIG. 1, the embodiment of the present application provides a smart rehabilitation device, The system comprises a mobile cart 1, an embedded host computer 11 and a mobile unit 12 each attached to the mobile cart 1, and further comprises a wearable unit 2, with the mobile unit 12 and the wearable unit 2 both connected to the embedded host computer 11.

[0022] The embedded host computer 11 includes a processor and a memory. The processor is for executing programs of a data collection module 111, a data processing module 112, a movement control module 113, a rehabilitation training module 114, and a rehabilitation evaluation module 115.

[0023] The processor may be a general-purpose processor (CPU, central processing unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits for executing the programs of the present application.

[0024] The memory may be the storage module 116 of the present application, and may be, but is not limited to, a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, a random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, an electrically erasable programmable read-only memory (EEPROM), a magnetic disk storage medium or other magnetic storage device, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer.

[0025] 2 , the wearable unit 2 includes rehabilitation training gloves 201 and a sensory device 202, both of which perform rehabilitation training via a rehabilitation training module 114 and transmit rehabilitation training data to a data collection module 111. The wearable unit 2 realizes wireless communication with the embedded host computer 11 using ZigBee, Bluetooth (registered trademark), or Wi-Fi wireless communication protocols, thereby enabling rehabilitation training to be performed based on the training content of the rehabilitation training module 114 and rehabilitation training data to be transmitted to the data collection module 111.

[0026] The data collection module 111 is for collecting information on the rehabilitation training patient, and the information includes eigenvectors corresponding to multiple dimensions of the patient, each eigenvector representing patient information in each dimension, and the dimensions include the patient's medical history file, lifestyle habits, environmental factors, and rehabilitation training status.

[0027] A patient's medical history file includes the patient's chief complaint, current medical history, past medical history, personal life history, marital and reproductive history, family history, diagnostic records, treatment recommendations, surgical records, pathology test reports, etc. Lifestyle habits include diet, exercise, smoking history, alcohol consumption, and work breaks. Environmental factors include temperature, humidity, air dust, wind force, environmental noise decibels, and air quality. Rehabilitation training includes motor function, balance function, and cognitive function. Naturally, in addition to these dimensional factors that affect a patient's rehabilitation, other factors, such as the patient's gender, address, age, education level, family situation, marital and reproductive history, economic situation, occupation, job type, and work environment, also affect the patient's rehabilitation. Taking lifestyle habits as an example, the eigenvector of the lifestyle dimension is established as (0, 0, 1, 1, 0, 0, etc.), including eating and drinking habits (normal: 0, saltier: 1, thinner: 2, excessive oil intake: 3, etc.), exercise status (moderate exercise: 0, no exercise: 1, small amount of exercise: 2, large amount of exercise: 3, excessive exercise: 4, etc.), smoking status, drinking status, and whether work breaks are regular or not.

[0028] The data processing module 112 includes a first prediction unit, a processing unit, a second prediction unit, and a generation unit, wherein the first prediction unit is for predicting the impact degree of each dimension of patient information on the patient's rehabilitation based on the eigenvector of each dimension to obtain a first prediction vector corresponding to the eigenvector of each dimension, and the first prediction vector is for indicating the impact degree of each dimension of patient information on the patient's rehabilitation; the processing unit is for feature combining each of the first prediction vectors to obtain a target vector; the second prediction unit is for predicting the rehabilitation effect corresponding to each dimension of patient information based on the target vector to obtain a second prediction vector, and the second prediction vector is for indicating the rehabilitation effect corresponding to the patient information of each dimension; and the generation unit is for generating prescription information for the patient's rehabilitation based on the second prediction vector, and the prescription information includes medication information and drug dosage.

[0029] The prescription information includes information such as the name of the medical institution, the patient's name, sex, age, outpatient or inpatient medical record number, department or ward and bed number, clinical diagnosis, and preparation date, as well as information such as the drug name, dosage form, specifications, number, and usage / dosage.

[0030] As an option, the first prediction unit specifically inputs an eigenvector corresponding to each dimension into a first prediction model corresponding to each dimension to obtain a first prediction vector corresponding to each eigenvector, and the first prediction vector indicates the impact of the patient information of each dimension on the patient's rehabilitation, and each of the first prediction models is an LSTM-based neural network model, and each of the LSTM-based neural network models is a model trained using a plurality of patient information samples and corresponding tag information.

[0031] Specifically, the second prediction unit inputs the target vector into a predetermined second prediction model to obtain a second prediction vector corresponding to each of the first prediction vectors, and the second prediction vector indicates a predicted result of rehabilitation effect corresponding to each dimension of patient information, and the second prediction model is an attention mechanism-based neural network model, which is a model trained using multiple samples of impact information on patient rehabilitation of each dimension of patient information and corresponding tag information.

[0032] The processing unit is further used for obtaining an activation function, the activation function including a weighting coefficient for each dimension and a predetermined bias constant; inputting the second predicted vector into the activation function; and determining at least one target sub-vector from the second predicted vector using the activation function. The generating unit is further for generating prescription information for rehabilitation of said patient based on said at least one target sub-vector.

[0033] The processing unit is further used for: acquiring a rehabilitation prescription pattern database, wherein the rehabilitation prescription pattern database includes a plurality of prescription information; generating each prescription information in the rehabilitation prescription pattern database as a corresponding prescription vector; acquiring a matching function and a weight vector, wherein the matching function is for matching the corresponding prescription vector based on the target sub-vector and the weight vector; and constructing a prescription generation model based on the prescription vector, the matching function, and the weight vector, wherein the weight vector is a predetermined parameter of the model.

[0034] The processing unit is further used for obtaining predicted prescription information generated for a plurality of patients using a prescription generation model, and obtaining actual prescription information actually administered corresponding to each predicted prescription information, wherein the actual prescription information includes actual medication information of the patient; obtaining a loss function of the prescription generation model; and training the prescription generation model using each predicted prescription information and the corresponding actual prescription information until a threshold of the loss function of the prescription generation model satisfies a predetermined condition, thereby obtaining a target prescription generation model.

[0035] The generation unit is specifically for inputting the at least one target sub-vector into the target prescription generation model to obtain prescription information for the patient's rehabilitation.

[0036] Specifically, the second prediction unit may optionally be selected to output a prediction of the influence of lifestyle habits, environmental factors, rehabilitation training, etc. on rehabilitation effect according to a multi-output control policy, and the multi-output control policy may include: JPEG2025531904000002.jpg5170 where σ is the activation function, which is initially set as the sigmod function or another function and can be selected according to actual needs, x is the second prediction vector, W and V are weighting coefficients, and b and c are bias constants.

[0037] In addition, the loss function of the prescription generation model is JPEG2025531904000003.jpg28170

[0038] Furthermore, the second prediction unit may further use a reward and punishment system and an error feedback method in the model training process to improve the model training efficiency. JPEG2025531904000004.jpg11170 where i and j respectively represent the dimension numbers, N(i) is the predicted prescription information vector, N(j) is the actual prescription information vector, and l is the reward / punishment constant.

[0039] The error feedback method is JPEG2025531904000005.jpg11170 However, observed t is the actual prescription information, and predicted t is the predicted prescription information, RMSE indicates the root mean square error of the system, N is the eigenvector dimension, and t is the prescription vector number.

[0040] In order to improve the visibility of the patient rehabilitation performance management, the processing unit may further monitor the patient's performance through compliance calculations. JPEG2025531904000006.jpg6170, where Task indicates the predicted prescription information vector, Execute indicates the actual prescription information vector, and Compliance i indicates the desired performance of the i-th predicted prescription information, and n is the number of the information vector.

[0041] JPEG2025531904000007.jpg12170 However, w i indicates the weight vector of the i-th predicted prescription information. Finally, the patient's performance pattern and the type of composite factor overall calculation value are output.

[0042] In an embodiment of the present application, the first prediction unit, the processing unit, the second prediction unit, and the generation unit may each be one or more processors, controllers, or chips having a communication interface and capable of implementing a communication protocol, and may also include memory and related interfaces, system transmission buses, etc., if necessary, and the processors, controllers, or chips implement corresponding functions by executing program code. Alternatively, an alternative solution is that the first prediction unit, the processing unit, the second prediction unit, and the generation unit share a single integrated chip or share devices such as a processor, controller, and memory. The shared processor, controller, or chip implements corresponding functions by executing program code.

[0043] The movement control module 113 obtains the position of the rehabilitation training patient, measures the distance to the rehabilitation training patient, and controls the mobile carriage 1 to move to the rehabilitation training patient, and adjusts the height to suit rehabilitation training patients of different heights, thereby obtaining a better user experience and providing great convenience for patients who are difficult to move.

[0044] The rehabilitation training module 114 is for providing a scenario training game designed based on rehabilitation movements, and the scenario training game is for realizing the training of the upper limbs, lower limbs and hands of a rehabilitation training patient.

[0045] The rehabilitation evaluation module 115 is for evaluating the motor functions of the upper limbs, lower limbs, and hands of a rehabilitation training patient. After the rehabilitation training module 114 completes the rehabilitation training, the rehabilitation evaluation module 115 performs an evaluation in the form of a score based on the training results.

[0046] Optionally, as shown in Fig. 3, the rehabilitation assessment module 115 includes an assessment auxiliary unit and an assessment unit. The assessment auxiliary unit is for completing the task of identifying rehabilitation assessment movements. The assessment auxiliary unit is a two-part cascade model, which includes a first classification subunit and a second classification subunit, respectively. The assessment unit is composed of a smart scale assessment subunit and a compensatory movement quantification assessment subunit.

[0047] The first classification subunit uses a neural network structure to identify whether the rehabilitation evaluation movement being performed by the patient is an upper limb arm movement, a palm movement, a lower limb movement, or an upper limb movement. The neural network's input is a movement eigenvector (a vector obtained by thresholding the original signal from the wearable unit), and its output is one of the four movement categories. The neural network includes one input layer, and the number of neurons in this layer is the length of the movement eigenvector, which is preset as N. The network includes two hidden layers, with the number of neurons in the first hidden layer being 2N and the number of neurons in the second hidden layer being N. The network includes one output layer, with the number of neurons being 4. Finally, the network completes the final output through a softmax layer and one-hot coding. The network calculation method is as follows: JPEG2025531904000008.jpg35170However, a j is the output of the hidden layer, σ is the activation function, N is the number of neurons, i is the loop parameter, and W ij are the weights and X i is the input and b j is the bias parameter, and z k is the kth output of the output layer, and W jk is the weight and b k is the bias parameter, output is the final output, ont-hot is the coding function, and softmax(z) is the activation function.

[0048] The second classification subunit uses a gated recurrent network, whose role is to identify specific rehabilitation evaluation actions. The input is the original signal of the rehabilitation wearable device, and the output is the specific identification result. The calculation method is as follows: JPEG2025531904000009.jpg32170, W r , W z and W h is the weight parameter, and b r , b h and bz is the bias parameter, which is automatically optimized by the backpropagation algorithm, and h t is the state at timestamp t, and h t-1 is the state at timestamp t-1, and h t ~ is the state of the updated timestamp, and g z is the control vector of the reset gate, and g r is the control vector of the update gate, σ is the activation function, commonly used as a Sigmoid function, and tanh is the tanh activation function.

[0049] The smart scale evaluation subunit is for completing the quantification score of rehabilitation movements. This subunit uses the signal from the wearable unit as input and outputs the smart quantification score of the movement. This part is completed using a self-attention mechanism-based LSTM network. First, attention coding is performed on the multi-dimensional signals (x1-xn) from the wearable unit using the self-attention mechanism. The calculation of the attention mechanism module is as follows: JPEG2025531904000010.jpg43170However, W q is the query matrix, and W k is the key matrix, which is automatically learned during the network training backpropagation process, and x i and x j are the input eigenvectors at different time points, AS is the attention score, AW is the attention weight, and x j ′ is the reconstructed eigenvector at a single time point, i and k are loop parameters, exp is the base e exponential function, and softmax is the activation function.

[0050] After self-attention encoding, the original signal (x1-x n ) is the reconstructed signal (x1′-x n '), and then rewrite it as (z1-z n) Then, the reconstructed signal is input to the LSTM network to complete the classification task. Assuming that the score of a single behavior in the original clinical scale is k-point system (0, 1...k), the LSTM completes the k-classification task. The calculation of the LSTM is as follows: JPEG2025531904000011.jpg40170However, the input gate is cell state C t (C t-1 is the cell state at the previous timestamp) and decides which new information to store in the cell state. The input gate consists of two parts. For the input part, it constructs the input information and decides which information to add to the unit state as new internal memory. For the candidate state part, it constructs one candidate cell state C t ~ After constructing these two parts, the unit state C t The output gate updates the final output h t (h t-1 is the hidden state at the previous timestamp). The output is the new unit state C t Based on this, it is completed in two steps. The first step is to decide which part of the cell state needs to be output. The second step is to control the final output. f , W i and W o is the weight parameter, and b f , b i , b c and b o is the bias parameter, which is automatically optimized by the backpropagation algorithm, and x t is the input eigenvector at time t, and i t is the input gate, and o t is the output gate, σ is the activation function, commonly used as a Sigmoid function, and tanh is the tanh activation function.

[0051] The output of the LSTM is passed through a softmax layer, which gives the final probability output (p1-p k) can be obtained, and then through one-hot coding, the classification output, i.e., the final smart score, can be completed. The smart scale evaluation subunit not only provides the smart score of the scale, but also provides the scale score s after subdivision. Its calculation method (i is the loop parameter) is as follows: The compensatory movement quantification assessment subunit includes movement compensation detection and compensatory movement quantification. The compensatory movement detection model uses an internal classifier to determine the probability that the movement parameter features will be identified as different types of compensatory movement (seven types in total), and the highest probability value and the corresponding name of the compensatory movement are used as the detection result. If the detection result shows that compensation is present, the compensatory movement quantification assessment model uses the quantification threshold range of the severity of the compensatory movement as the evaluation tag of the quantification category, determines a quantification threshold range corresponding to the movement parameter features based on the K-nearest neighbor algorithm, and outputs the quantification category corresponding to the quantification threshold range as the compensatory movement quantification assessment result.

[0052] The internal classifier for compensatory movement detection is a support vector machine model. First, feature extraction is performed on the limb movement data based on the completeness, smoothness, and variability of the target patient's limb movement test movements. Movement parameter features related to the target patient's trunk, shoulder joint, and elbow joint are selected to obtain movement eigenvectors. To detect the seven types of compensatory movements present in the target patient, seven binary classification models are constructed to represent the compensatory movements in the corresponding categories. Finally, the output results of the seven SVM binary classifiers are combined to detect and classify multiple types of compensatory movements. That is, each binary classification model takes an eigenvector as input and outputs probabilities p0 and p1 for classifying the movement as 0 or 1. If p1 can be greater than p0, the largest of the seven models' p1 outputs is the final compensatory movement type. If p1 is never greater than p0, no compensation exists.

[0053] The compensatory movement quantification evaluation model uses the quantification threshold interval of the severity of compensatory movement as an evaluation tag of the quantification category, determines the quantification threshold interval corresponding to the movement parameter feature based on the K-nearest neighbor algorithm, and outputs the quantification category corresponding to the quantification threshold interval as the compensatory movement quantification evaluation result.

[0054] Specifically, the quantification threshold intervals for the severity of compensatory movement are determined by dividing the same compensatory movement into K levels according to its severity, and the division ranges for each severity level are the quantification threshold intervals for compensatory movement. The quantification threshold intervals for mild, moderate, and severe compensation are determined by performing clustering analysis on multiple movement parameter features of stroke patients with such compensatory movement using a K-clustering algorithm. Clustering analysis is performed on large sample data using a K-means clustering algorithm to obtain tags for the division intervals of compensatory movement severity, and K-level compensatory movement severity assessment thresholds are obtained by parameter setting, where K is 3 or greater. When K is equal to 3, the quantification threshold intervals corresponding to mild, moderate, and severe compensation are the first, second, and third threshold intervals, respectively. Then, a supervised machine learning algorithm is used to separate the training set and test set, and the movement parameter features and the multiple-level assessment tags are used as inputs for the compensatory movement quantification assessment model. A K-nearest neighbor machine learning algorithm is used to realize a quantitative assessment of the severity of compensatory movement based on the multiple-level threshold tags.

[0055] In an embodiment of the present application, the first prediction unit, the processing unit, the second prediction unit, the generation unit, the evaluation auxiliary unit, the evaluation unit, the smart scale evaluation subunit, the compensatory movement quantification evaluation subunit, the first classification subunit, and the second classification subunit may each be one or more processors, controllers, or chips having a communication interface and capable of implementing a communication protocol, and may further include memory and related interfaces, a system transmission bus, etc., if necessary, and the processor, controller, or chip executes program-related code to realize corresponding functions. Alternatively, an alternative solution is that the first prediction unit, the processing unit, the second prediction unit, the generation unit, the evaluation auxiliary unit, the evaluation unit, the smart scale evaluation subunit, the compensatory movement quantification evaluation subunit, the first classification subunit, and the second classification subunit share an integrated chip or share devices such as a processor, controller, memory, etc. The shared processor, controller, or chip executes program-related code to realize corresponding functions.

[0056] 4, the moving unit 12 includes rollers 123, an electric push rod 122, and a driving device 121 installed on the moving cart 1, and the driving device 121 is connected to the moving control module 113. The moving control module 113 includes a controller, which controls the driving device 121 to drive the rollers 123 to move, and controls the electric push rod 122 to extend and retract, thereby realizing height adjustment for the embedded host computer 11 so that the height can be adapted to the current rehabilitation training patient.

[0057] In this embodiment, the embedded host computer 11 is an SK-27A all-in-one PC with a 27-inch capacitive touch panel, an Intel Core i5-3210 CPU, 4GB of Longsys memory, and a 128GB Longsys solid-state drive, and also includes a power adapter, connecting cables, and a stylus. The all-in-one PC is mounted on the top of the mobile carriage 1 in an embedded manner. The mobile carriage 1 further includes a base 101 and a connection block 102, as shown in FIG. 2. Rollers 123 are installed at the bottom of the base 101, and an electric push rod 122 is installed in the connection block 102. The embedded host computer 11 is connected to the output terminal of the electric push rod 122, thereby controlling the height of the embedded host computer 11. A TJC-C4 electric push rod is selected and used as the electric push rod.

[0058] It should be noted that in this embodiment, the rehabilitation training module 114 and the rehabilitation assessment module 115 constitute a rehabilitation training and assessment system. After logging in to the rehabilitation training and assessment system, rehabilitation training and rehabilitation assessment content can be selected as needed. After selecting the rehabilitation training content, the user first needs to put on the rehabilitation training gloves 201 and the haptic device 202 and then select the rehabilitation training function based on the display window. The rehabilitation training function is loaded with many rehabilitation training games, so the rehabilitation training patient can select the corresponding difficulty level of the rehabilitation training game according to their own situation. For patients using the device for the first time, the system also loads instruction videos, so the rehabilitation training patient can learn how to use the device based on the videos, helping the user to better use the device and further improving the user experience. The rehabilitation training games are selected from fun games that are relevant to daily life, and can fully train the patient's upper limbs, lower limbs, and fingers through the games, accelerating the patient's rehabilitation progress.

[0059] 5, rehabilitation training gloves 201 include a glove body and a first housing attached to the glove body, a first battery, a first circuit board, and a bending sensor installed in the first housing, and a first indicator lamp installed in the first housing. In this embodiment, the glove body is made of nylon material, and the opening of the glove is elastic to improve comfort. The gloves are available in several different styles and are suitable for many people. The first battery is a Zhongshun 602035 type lithium battery, which provides power support for the remaining components on the glove; the bending sensor is a FS-L-0055-253-ST bending sensor, which is used to measure relevant data when a rehabilitation training patient is training; the first circuit board is used to transmit the data from the bending sensor to the data collection module 111 via the integrated wireless transmission module; at the same time, the first circuit board is also used to control the flashing of a first indicator light; when the channel device corresponding to the rehabilitation training glove 201 is operating normally, the first circuit board controls the indicator light to be green (or flashing); when there is no data in the corresponding channel device, the first circuit board controls the indicator light to be gray; and when the data reception by the corresponding channel device is abnormal, the indicator light is controlled to be red.

[0060] As shown in FIG. 6 , the haptic device 202 includes a band and a second housing attached to the band. A second circuit board and a second battery are installed in the second housing, and a second indicator lamp is installed in the second housing. In this embodiment, the haptic device 202 may be multiple and is mainly used to collect upper and lower limb movement data. The band may be selected to have different sizes as needed and is used to secure the second housing and the second circuit board and second battery therein in corresponding positions. The second battery may be the same type as the first battery and is used to provide power support. A corresponding sensor, such as a posture sensor, is installed in the second housing and connected to the second circuit board as needed. The second circuit board controls the flashing of the second indicator lamp and also transmits sensor data to the data collection module 111 via an integrated wireless transmission module.

[0061] As will be understood, in this embodiment, in addition to the rehabilitation training gloves 201 and the sensory device 202, other wearable devices or rehabilitation training devices, such as a walking machine, a rehabilitation training helmet, etc. may also be used.

[0062] Preferably, a receiver 1111 and a camera 1112 are connected to the data collection module 111, and the receiver 1111 is both wirelessly connected to the rehabilitation training gloves 201 and the haptic device 202 for receiving motion data from devices such as the rehabilitation training gloves 201 and the haptic device 202, and the camera 1112 captures the movements of the rehabilitation training patient to collect the motion data. In this embodiment, the camera 1112 may be a camera 1112 attached to the all-in-one computer or a separately connected camera 1112; the type of camera 1112 is not particularly required as long as it is compatible with the all-in-one computer and the rehabilitation training and evaluation system.

[0063] Preferably, in this embodiment, the all-in-one computer can be connected to a cloud server via a network, where the exercise prescription is stored, and the exercise prescription can be downloaded or training data can be uploaded through the cloud server. In practical application, rehabilitation training patients can also be connected to terminal devices such as hospitals through the cloud server, allowing doctors to provide remote online guidance and diagnosis.

[0064] As shown in Figure 4, the device further includes a man-machine interactive support 13, which is connected to the all-in-one computer via a wired or wireless connection. The man-machine interactive support 13 is provided with a card reading area and a button area. The card reading area is for identifying the rehabilitation patient, and the button area is provided with a number of buttons for assisting the progression of the scenario training game. Before using the device for rehabilitation training, the rehabilitation patient first swipes their ID card or ID card in the card reading area. The all-in-one computer then authenticates the patient using the ID card or ID card. The rehabilitation patient then enters the rehabilitation training and evaluation system and can play the rehabilitation training game by pressing the buttons in the button area. As can be seen, a rehabilitation patient management system can be established based on the man-machine interactive support 13, and files can be established for each rehabilitation training and evaluation session of the patient. This not only allows for real-time tracking of the patient's rehabilitation progress and improves the user experience, but also reduces the burden on operators and standardizes the entire rehabilitation training process through detailed management files.

[0065] Preferably, the device further includes a power supply module 14, which includes a magnetic core, a transformer, a filter, and data lines, and is used to supply power and process power signals. The magnetic core is selected from V18004, V18005, or V18007 types to provide electromagnetic shielding. The filter is selected from RSEN-2006L EMC filters to process power signals. The transformer is selected from H0128-823-0250 type isolation transformers. The magnetic core, transformer, and filter are all installed in an empty cavity within the base 101, which increases the weight of the base 101 and helps maintain the stability of the connected embedded host computer 11 during movement. Each component is connected to a power supply and the embedded host computer 11 as needed, and detailed descriptions are omitted here.

[0066] 7, a radar detector is connected to the mobile control module 113, the radar detector is for detecting the position of the rehabilitation training patient, and the mobile control module 113 is for controlling the driving device 121 and the electric push rod 122 to operate. In this embodiment, the radar detector may be selected to be a SENKYLASER SK-C10 2D laser scanning radar, and the mobile control module 113 realizes autonomous positioning and map construction using a Gmapping algorithm. It should be understood that the device further includes a GPS positioning module to determine the current position of the mobile carriage 1, and then scans the surrounding environment using the radar detector to determine the position of the rehabilitation training patient, and finally the mobile control module 113 controls the driving device 121 to move the rollers 123 toward the rehabilitation training patient. In this embodiment, there are four rollers 123, which are arranged circumferentially on the bottom of the mobile carriage 1. The driving device 121 selectively uses two motors, one of which is connected to two rollers 123. The forward and reverse rotation of the motor is controlled by a controller to drive and rotate the rollers 123, thereby moving the mobile carriage 1 forward and backward. The other motor is connected to the other two rollers 123. The forward and reverse rotation of the other motor is controlled by a controller to drive and rotate the other two rollers 123, thereby allowing the mobile carriage 1 to turn left and right. The method of controlling the forward, backward, left, and right movement of the device using a motor is conventional and similar to the method of controlling the forward, backward, left, and right movement of a toy car using a steering wheel. The controller may optionally use a remote control, which is connected to the motor via Wi-Fi and has forward, backward, left, and right buttons on the remote control, allowing the forward, backward, left, and right movement of the device to be controlled by the remote control.

[0067] It should be noted that the drive unit 121 uses a servo motor as a drive unit, and its type may be the HC-KFS, HC-MFS, HC-SFS, HC-RFS, or HC-UFS series. The output terminal of the drive unit 121 is connected to the rollers 123, and the movement control module 113 controls the drive unit 121, thereby realizing the movement of the carriage 1. In order to better realize the forward, backward, leftward, and rightward movement of the carriage 1, the number of rollers 123 is four and the number of drive units 121 is two. The four rollers 123 are installed at the bottom of the base 101, and two rollers 123 are connected to one drive unit 121, and the forward and reverse rotation of the drive unit 121 is controlled to realize the forward and reverse movement of the carriage 1, and the other two rollers 123 are connected to the other drive unit 121, and the forward and reverse rotation of the drive unit 121 is controlled to realize the left and right turning of the carriage 1.

[0068] Since different users have different heights, to realize automatic height adjustment of the device, an infrared sensor connected to the movement control module 113 is installed on the top of the moving carriage 1, and the patient only needs to wear a corresponding infrared signal transmitter when using the device. When starting to use, the movement control module 113 controls the electric push rod 122 to operate within a certain range, moving the embedded host computer 11 within a certain height range. When the infrared sensor and the infrared signal transmitter are at the same height, the infrared sensor will receive the infrared signal transmitted from the infrared signal transmitter, at which time the movement control module 113 controls the electric push rod 122 to stop operating, thereby obtaining the height most suitable for the current patient.

[0069] In another embodiment, the device can be manually adjusted for height. The lifting / lowering program for the electric push rod 122 can be integrated into the movement control module 113, and the lifting / lowering control screen for the electric push rod 122 of the movement control module 113 can be opened and the lifting / lowering option can be clicked to lift or lower the embedded host computer 11.

[0070] In another embodiment, the movement control module 113 includes a remote control, which is connected to the embedded host computer 11 via WIFI, and a corresponding up or down button on the remote control realizes control over the motorized push rod 122 of the movement control module 113, thereby raising or lowering the embedded host computer 11.

[0071] In this embodiment, the data collection module 111 collects original limb movement signals of the rehabilitation training patient using a receiver 1111 and a camera 1112, and transmits them to the data processing module 112 for processing. The data processing module 112 must first preprocess the limb movement signals to remove interference and influence of system noise, and then extract features from the original limb movement signals. In this embodiment, the feature extraction is performed in terms of the completeness and smoothness of the patient's movement completion, to obtain one-dimensional eigenvectors of the rehabilitation training movements. Finally, a multi-source fusion algorithm is used to quantify and evaluate the multi-dimensional features, and based on the results, driving (control) and man-machine interactive feedback is provided for tasks and characters in the virtual reality scene.

[0072] The rehabilitation evaluation module 115 establishes a motion discrimination model for the wearable device based on a sequence network, and after the patient completes the rehabilitation training motion, the patient needs to evaluate the quality of the training motion. After obtaining the one-dimensional eigenvector of the rehabilitation training motion processed by the data collection module 111, an attention-based LSTM scale evaluation network and a machine learning-based compensation detection network are used to establish a training motion evaluation model. During motion evaluation, the process is predicted, i.e., evaluated (giving a motion score and a compensation score), and the data is incorporated into a training set to update the training model, allowing the model to have self-learning capabilities. Based on the model, nonlinear dynamics tools such as Poincare difference scatter plots, maximized Lyapunov exponent, and distribution entropy are used to design visualized quantification evaluation parameters, which enable the quantitative evaluation model to provide a visualized quantification score for the training motion.

[0073] The above description is merely the most preferred specific embodiment of the present application, but the scope of protection of the present application is not limited thereto, and any changes or substitutions within the technical scope disclosed in the present application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be governed by the claims. [Explanation of symbols]

[0074] 1 Mobile cart 101 Pedestal 102 Connection Block 11 Embedded Host Computers 111 Data Collection Module 1111 receiver 1112 Camera 112 Data Processing Module 113 Movement Control Module 114 Rehabilitation Training Module 115 Rehabilitation Assessment Module 116 Memory Module 12 Mobile Units 121 Drive unit 122 Electric push rod 123 Laura 2 Wearable unit 201 Rehabilitation training gloves 202 Sensory Device 13 Man-machine interactive support 14 Power Supply Modules

Claims

1. A smart rehabilitation device, the device comprising: a mobile carriage; an embedded host computer, a mobile unit, and a wearable unit are installed on the mobile carriage, and the mobile unit and the wearable unit are both communicatively connected to the embedded host computer; the embedded host computer includes a rehabilitation training module, a rehabilitation assessment module, a data collection module, a data processing module, a movement control module, and a storage module; The rehabilitation training module is for providing a scenario training game designed based on rehabilitation movements, and the scenario training game is for training the upper limbs, lower limbs and hands of a rehabilitation training patient; the rehabilitation evaluation module is for evaluating the motor functions of the upper limbs, lower limbs, and hands of the rehabilitation training patient; The data collection module is for collecting information about the rehabilitation training patient, the information including eigenvectors corresponding to multiple dimensions of the patient, each eigenvector representing patient information in each dimension, the dimensions including the patient's medical history file, lifestyle habits, environmental factors, and rehabilitation training status; the data processing module is used for: predicting the degree of impact of each dimension of patient information on the patient's rehabilitation based on the eigenvector of each dimension, to obtain a first prediction vector corresponding to the eigenvector of each dimension, the first prediction vector being for indicating the degree of impact of each dimension of patient information on the patient's rehabilitation, and feature combining each of the first prediction vectors to obtain a target vector; predicting the rehabilitation effect corresponding to each dimension of patient information based on the target vector, to obtain a second prediction vector, the second prediction vector being for indicating the rehabilitation effect corresponding to the patient information of each dimension; and generating prescription information for the patient's rehabilitation based on the second prediction vector, the prescription information including medication information and drug dosage; the movement control module acquires the position of the rehabilitation training patient and controls the moving carriage to move its position and adjust its height by measuring the distance between the rehabilitation training patient and the moving carriage; The moving unit includes a roller, an electric push rod, and a driving device installed on the moving carriage, the driving device being connected to the movement control module, and the movement control module driving the roller by the driving device to move and adjust the position, and controlling the electric push rod to adjust the height; The wearable unit includes rehabilitation training gloves and a sensory device, and both the rehabilitation training gloves and the sensory device perform rehabilitation training using the rehabilitation training module and transmit rehabilitation training data to the data collection module.

2. the data collection module includes a receiver and a camera, the receiver being wirelessly connected to the rehabilitation training glove and the haptic device, respectively; the receiver is for receiving motion data of the rehabilitation training gloves and the sensory device, and the camera captures the motion of the rehabilitation training patient to collect the motion data; 10. The apparatus of claim 1, wherein the data processing module is further for processing the motion data by multi-source information fusion.

3. The apparatus further comprises: a man-machine interactive support having a card reading area and a button area, the card reading area being for identifying the rehabilitation training patient, and a plurality of buttons being provided in the button area, the buttons being for assisting the progression of the scenario training game; a power supply module including a magnetic core, a transformer, a filter and a data line, for supplying power and processing a power signal; the embedded host computer is further communicatively connected to a cloud server; The exercise prescription is stored in the cloud server, 2. The device of claim 1, wherein the embedded host computer is further for downloading the exercise prescription or uploading the rehabilitation training data by the cloud server.

4. The rehabilitation training glove includes a glove body and a first housing attached to the glove body, 2. The device of claim 1, wherein a first battery, a first circuit board, and a bending sensor are disposed within the first housing, and a first indicator lamp is disposed on the first housing.

5. The haptic device includes a band and a second housing attached to the band, 2. The device of claim 1, wherein a second circuit board and a second battery are disposed within the second housing, and a second indicator light is disposed within the second housing.

6. The rehabilitation evaluation module includes an evaluation auxiliary unit and an evaluation unit, and the evaluation unit includes a smart scale evaluation subunit and a compensatory movement quantification evaluation subunit; The evaluation assistance unit is for completing a rehabilitation evaluation movement identification task and obtaining an identification result; The apparatus according to claim 1 , wherein the evaluation unit is for obtaining a corresponding evaluation result based on the identification result.

7. 2. The device according to claim 1, wherein the embedded host computer is used for remote medical diagnosis or online rehabilitation training guidance, realizing interconnection of terminals via a network.

8. the movement control module is connected to a radar detector and a positioning module; the radar detector is for detecting the position of the rehabilitation training patient; 2. The apparatus of claim 1, wherein the positioning module is for determining a current position of the mobile carriage.

9. the moving carriage includes a base and a connection block; The device of any one of claims 1 to 8, characterized in that the roller is installed at the bottom of the base, the electric push rod is installed in the connection block, and the embedded host computer is connected to the top of the electric push rod.

10. The device of claim 1, wherein an infrared sensor and an infrared signal transmitter are connected to the movement control module, and the device realizes position and height adjustments by the infrared sensor, the infrared signal transmitter and the movement control module when in use.

Images