Active hand function rehabilitation training system and method

Abstract

The application discloses an active hand function rehabilitation training system and method, and relates to the technical field of rehabilitation medicine. The system comprises at least one interactive operation module and a master control device. The interactive operation module collects mechanical data formed by active operation of a user, obtains a current operation state according to the mechanical data and an operation target parameter, and outputs a guiding feedback. The master control device obtains a plurality of motion evaluation features representing active force generation capacity, force control stability and sustained control capacity according to the mechanical data, obtains a training evaluation result according to the plurality of motion evaluation features and a target evaluation feature, and updates a training target parameter of a next training round according to the training evaluation result. The application can quantitatively evaluate the active operation process in multiple dimensions, and adjust the guiding feedback and the training target parameter according to the performance of the user.

Description

Technical Field

[0001] This application relates to the field of rehabilitation medicine, and in particular to an active hand function rehabilitation training system and method. Background Technology

[0002] Hand function rehabilitation trainers are widely used in the rehabilitation process for patients with stroke, hand injuries, and other conditions that lead to hand motor dysfunction. Patients typically need to perform active training movements such as repeated finger pressing, strength maintenance, and coordination control to gradually restore hand muscle strength and motor control.

[0003] Current hand function rehabilitation training mainly relies on therapists to observe patients' movements and assess them using scales. Some electronic training devices record the number of times patients press, the time taken to complete the task, or whether the preset force is reached through pressure sensors or buttons.

[0004] While the methods described above can reflect whether a patient has completed the training movements, they struggle to continuously acquire data on changes in force, stability of force, maintenance of target force, and pressure distribution during the patient's active exertion. Therefore, they cannot accurately assess the quality of the patient's performance. Furthermore, existing equipment typically provides simple success or failure indicators upon completion of the movement, failing to offer timely and directional feedback based on the patient's current exertion level during the movement.

[0005] In addition, existing training equipment mostly uses fixed target force, holding time and allowable error, which cannot dynamically adjust subsequent training parameters according to the patient's training performance, ability improvement or fatigue state, which can easily lead to a mismatch between training difficulty and the patient's actual ability.

[0006] Therefore, providing an active rehabilitation training method that can collect multidimensional biomechanical information during a patient's active operation, objectively quantify the quality of the movement, provide guidance feedback based on the real-time operation status, and adaptively adjust subsequent training parameters is a technical problem that urgently needs to be solved by those in the field. Summary of the Invention

[0007] The purpose of this application is to provide an active hand function rehabilitation training system and method to solve the problems of existing hand function rehabilitation training equipment having a single evaluation dimension, delayed feedback, and fixed training parameters.

[0008] To address the aforementioned technical problems, this application provides an active hand function rehabilitation training system, comprising: at least one interactive operation module and a main control device; the interactive operation module includes a biomechanical information acquisition unit, a processing unit, and a feedback unit, wherein the processing unit is connected to the biomechanical information acquisition unit and the feedback unit respectively, and the main control device is communicatively connected to the interactive operation module; The main control device is used to acquire the current training target parameters and send the current training target parameters to the interactive operation module; the current training target parameters include operation target parameters and target evaluation features corresponding to multiple action evaluation features respectively; The mechanical information acquisition unit is used to collect mechanical data generated when the user actively operates the interactive operation module; The processing unit is used to obtain the current operation state based on the relationship between the currently collected mechanical data and the operation target parameters, and to control the feedback unit to output corresponding guidance feedback based on the current operation state; The main control device is used to acquire the mechanical data, obtain the plurality of action evaluation features based on the mechanical data, obtain training evaluation results based on the relationship between the plurality of action evaluation features and the corresponding target evaluation features, update the current training target parameters based on the training evaluation results, obtain the training target parameters corresponding to the next training round, and send the training target parameters corresponding to the next training round to the interactive operation module; wherein, the plurality of action evaluation features include at least a first action evaluation feature representing active force exertion ability, a second action evaluation feature representing force control stability, and a third action evaluation feature representing continuous control ability.

[0009] Furthermore, the mechanical information acquisition unit includes a pressure sensor array, which includes multiple pressure acquisition areas; The main control device is also used to obtain at least one spatial motion evaluation feature from the pressure distribution center, pressure action area and pressure distribution center trajectory based on the pressure data corresponding to each pressure acquisition area, and to use the spatial motion evaluation feature as one of the plurality of motion evaluation features.

[0010] Furthermore, the operational target parameters include the target force and the allowable force range, with the allowable force range including a lower limit and an upper limit. Based on the relationship between the current force and the lower and upper limits, the processing unit determines the states of insufficient force, adequate force, and excessive force, and controls the feedback unit to output differentiated guiding feedback.

[0011] Furthermore, the operational target parameters include target force, allowable force range, and target holding time; The main control device is used to obtain the first action evaluation feature based on the relationship between the peak force and the target force during an active operation. The main control device is also used to determine the first moment when the mechanical data first enters the force allowable range and the second moment when the mechanical data first leaves the force allowable range after the first moment, and to take the time period between the first moment and the second moment as a continuous holding period; and to obtain the second action evaluation feature based on the relationship between the force mean and force standard deviation of the mechanical data in the continuous holding period, wherein the second action evaluation feature increases as the force standard deviation decreases relative to the force mean. The main control device is also used to obtain the continuous holding duration based on the first time and the second time, and to obtain the third action evaluation feature based on the relationship between the continuous holding duration and the target holding duration.

[0012] Furthermore, the main control device is used to normalize the first action evaluation feature, the second action evaluation feature, and the third action evaluation feature respectively, and to perform weighted processing on the normalized first action evaluation feature, the second action evaluation feature, and the third action evaluation feature according to preset weights to obtain the training evaluation result; The main control device is also used to obtain the action evaluation features to be improved based on the relationship between the normalized first action evaluation features, the second action evaluation features and the third action evaluation features and their respective corresponding evaluation thresholds.

[0013] Furthermore, the main control device is used to acquire the training evaluation results corresponding to multiple consecutive training rounds, and obtain a stage training evaluation result based on the average value of the training evaluation results corresponding to multiple consecutive training rounds; when the stage training evaluation result is greater than or equal to the upper limit evaluation threshold, the main control device increases the target intensity, reduces the allowable range of intensity, or extends the target holding time; when the stage training evaluation result is greater than or equal to the lower limit evaluation threshold and less than the upper limit evaluation threshold, the main control device keeps the target intensity, the allowable range of intensity, and the target holding time unchanged.

[0014] Furthermore, the main control device stores the correspondence between action evaluation features, interactive tasks, and task control parameters; The main control device is used to determine the target interaction task and the task control parameters of the target interaction task based on the evaluation features of the action to be improved and the correspondence, update the task control parameters of the target interaction task based on the mechanical data, and control the operation of the target interaction task according to the updated task control parameters. The main control device is also used to adjust the task control parameters for the next training round or change the target interactive task based on the training evaluation results.

[0015] This application also provides an active hand function rehabilitation training method, applied to the above-mentioned active hand function rehabilitation training system, the method comprising: Obtain the current training target parameters and send the current training target parameters to the interactive operation module; the current training target parameters include operation target parameters and target evaluation features corresponding to multiple action evaluation features respectively; Acquire mechanical data generated when the user actively operates the interactive module; The interactive operation module obtains the current operation state based on the relationship between the currently collected mechanical data and the operation target parameters, and outputs corresponding guidance feedback based on the current operation state; The plurality of motion evaluation features are obtained based on the mechanical data. The plurality of motion evaluation features include at least a first motion evaluation feature characterizing active force exertion capability, a second motion evaluation feature characterizing force control stability, and a third motion evaluation feature characterizing continuous control capability. The training evaluation result is obtained based on the relationship between the multiple action evaluation features and their corresponding target evaluation features; The current training objective parameters are updated based on the training evaluation results to obtain the training objective parameters for the next training round, and the training objective parameters for the next training round are sent to the interactive operation module.

[0016] Furthermore, the current training target parameters include target intensity, allowable intensity range, and target holding duration; The process of obtaining the multiple motion evaluation features based on the mechanical data includes: The first action evaluation feature is obtained based on the relationship between the peak force and the target force during an active operation. Determine the first moment when the mechanical data first enters the force allowable range and the second moment when the mechanical data first leaves the force allowable range after the first moment, and take the time period between the first moment and the second moment as the continuous holding period; The second motion evaluation feature is obtained based on the relationship between the mean force and the standard deviation of the force of the mechanical data during the continuous holding period. The second motion evaluation feature increases as the standard deviation of force decreases relative to the mean force. The continuous holding duration is obtained based on the first time point and the second time point, and the third action evaluation feature is obtained based on the relationship between the continuous holding duration and the target holding duration; The first action evaluation feature, the second action evaluation feature, and the third action evaluation feature are normalized respectively, and then the normalized first action evaluation feature, the second action evaluation feature, and the third action evaluation feature are weighted according to preset weights to obtain the training evaluation result.

[0017] Furthermore, updating the current training target parameters based on the training evaluation results includes: Obtain the training evaluation results corresponding to multiple consecutive training rounds, and obtain the stage training evaluation results based on the training evaluation results corresponding to multiple consecutive training rounds. When the training evaluation result of the aforementioned stage is greater than or equal to the upper limit evaluation threshold, the target intensity is increased, the allowable range of the intensity is narrowed, or the target retention time is extended. When the training evaluation result of the aforementioned stage is less than the lower limit evaluation threshold, the target intensity is reduced, the allowable range of the intensity is expanded, or the target retention time is shortened. The current training target parameters are updated based on the adjusted target intensity, the allowable range of intensity, or the target duration, to obtain the training target parameters corresponding to the next training round.

[0018] The active hand function rehabilitation training system provided in this application collects mechanical data generated during the user's active operation of the interactive operation module through a mechanical information acquisition unit. Based on this mechanical data, multiple action evaluation features characterizing active force exertion ability, force control stability, and sustained control ability are obtained. This allows training evaluation to move beyond being limited to the number of movements, completion time, or a single force, and to reflect the quality of the user's active operation process from multiple aspects. By comparing multiple action evaluation features with corresponding target evaluation features, training evaluation results can be obtained based on a unified evaluation benchmark, achieving an objective evaluation of active rehabilitation training. The processing unit in the interactive operation module determines the current operation state based on the relationship between the currently collected mechanical data and the operation target parameters, and directly controls the feedback unit to output corresponding guidance feedback, enabling the user to promptly understand the current operation state and adjust the force exertion process during active operation. The main control device further updates the current training target parameters based on the training evaluation results, obtaining the training target parameters for the next training round, allowing subsequent training targets to be adjusted according to the user's training performance. This forms guidance feedback during the current movement process and target parameter updates between different training rounds, solving the problems of single evaluation dimensions, delayed feedback, and fixed training parameters in existing hand function rehabilitation training equipment.

[0019] In addition, this application also provides an active hand function rehabilitation training method, which corresponds to the above-mentioned active hand function rehabilitation training system and has the same effect. Attached Figure Description

[0020] To more clearly illustrate the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A structural block diagram of an active hand function rehabilitation training system provided in an embodiment of this application; Figure 2 A schematic diagram of force-time curve and evaluation parameters for single-finger pressing training provided in an embodiment of this application; Figure 3 A radar chart of three action evaluation features provided in an embodiment of this application; Figure 4 A flowchart of an active hand function rehabilitation training method provided in an embodiment of this application. Detailed Implementation

[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of this application.

[0023] The core of this application is to provide an active hand function rehabilitation training system and method.

[0024] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0025] like Figure 1 As shown, this embodiment provides an active hand function rehabilitation training system, including: at least one interactive operation module 110 and a main control device 120; the interactive operation module includes a biomechanical information acquisition unit 111, a processing unit 112 and a feedback unit 113, the processing unit is connected to the biomechanical information acquisition unit and the feedback unit respectively, and the main control device is communicatively connected to the interactive operation module; The main control device is used to acquire the current training target parameters and send the current training target parameters to the interactive operation module; the current training target parameters include operation target parameters and target evaluation features corresponding to multiple action evaluation features respectively; The mechanical information acquisition unit is used to collect mechanical data generated when the user actively operates the interactive operation module; The processing unit is used to obtain the current operation state based on the relationship between the currently collected mechanical data and the operation target parameters, and to control the feedback unit to output corresponding guidance feedback based on the current operation state; The main control device is used to acquire the mechanical data, obtain the plurality of action evaluation features based on the mechanical data, obtain training evaluation results based on the relationship between the plurality of action evaluation features and the corresponding target evaluation features, update the current training target parameters based on the training evaluation results, obtain the training target parameters corresponding to the next training round, and send the training target parameters corresponding to the next training round to the interactive operation module; wherein, the plurality of action evaluation features include at least a first action evaluation feature representing active force exertion ability, a second action evaluation feature representing force control stability, and a third action evaluation feature representing continuous control ability.

[0026] First, the system can be installed in hospital rehabilitation departments, community rehabilitation centers, or home training environments. During use, the user places their palm on the mounting plate and actively presses the interactive module 110 with one or more fingers. This active operation refers to the user's own hand muscle strength applied through pressing, holding, or releasing motions, and does not imply the application of therapeutic pressure to the fingers by an external device.

[0027] The active hand function rehabilitation training system includes at least one interactive operation module 110 and a main control device 120. The interactive operation module 110 is a component directly operated by the user and can manifest as buttons, pressure plates, grips, or other components capable of receiving finger forces. The main control device 120 can be a tablet computer, personal computer, embedded computing device, or server, or it can be implemented by a display terminal and a computing unit. This embodiment does not limit the physical form of the main control device 120, as long as it can receive mechanical data, generate movement evaluation features, and update training target parameters.

[0028] The interactive operation module 110 includes a force information acquisition unit 111, a processing unit 112, and a feedback unit 113. The force information acquisition unit 111 converts finger force into electrical signals or digital quantities. The processing unit 112 can be a microprocessor, microcontroller, or other electronic components with signal processing capabilities. The feedback unit 113 outputs visual, auditory, or tactile information to the user. Because the processing unit 112 is connected to both the force information acquisition unit 111 and the feedback unit 113, it can directly control the feedback unit 113 after obtaining the current force data, without waiting for the entire training session to end.

[0029] The main control device 120 is communicatively connected to the interactive operation module 110. This communication connection can utilize Bluetooth, wireless LAN, serial bus, universal serial bus, or other data transmission methods. If the system has multiple interactive operation modules 110, each interactive operation module 110 can establish a communication connection with the main control device 120, or it can first connect to the aggregation unit in the mounting board, and then the aggregation unit sends data to the main control device 120. This embodiment does not strictly limit the communication method.

[0030] The current training objective parameters are the set of parameters used in the current training round, including operational objective parameters and target evaluation features corresponding to multiple action evaluation features. Operational objective parameters are used for state judgment during the current action process and can be target force, allowable force range, target holding duration, pressing sequence, or operational rhythm. Target evaluation features are used for evaluation after training, and can be peak force targets, force stability targets, target holding rate targets, or spatial action evaluation targets. Operational objective parameters and target evaluation features can be partially the same, or they can be set separately according to their purpose.

[0031] At the start of training, the main control device 120 acquires the current training target parameters and sends them to the interactive operation module 110. This acquisition can be reading initial parameters input by the therapist, parameters from the user's history, or a pre-stored training plan. After receiving the current training target parameters, the interactive operation module 110 continuously collects the mechanical data generated during the user's active operation. The processing unit 112 compares the current mechanical data with the operation target parameters to obtain the current operation state and controls the feedback unit 113 to output corresponding guidance feedback.

[0032] Simultaneously, the main control device 120 acquires mechanical data and generates multiple motion evaluation features. The first motion evaluation feature characterizes the active force application capability and can be derived from peak force, average force, or the degree to which the target force is achieved. The second motion evaluation feature characterizes the stability of force control and can be derived from the standard deviation, coefficient of variation, or steady-state error during the holding phase. The third motion evaluation feature characterizes the continuous control capability and can be derived from the continuous holding duration, target holding degree, or number of interruptions during the holding phase. In other words, the three motion evaluation features evaluate the user's active operation quality from the perspectives of force application magnitude, force application stability, and continuous holding capability, respectively.

[0033] The main control device 120 compares multiple action evaluation features with their corresponding target evaluation features to obtain training evaluation results. These results can be a comprehensive score or individual scores for each action evaluation feature. The main control device 120 then updates the current training target parameters based on the training evaluation results, obtaining the training target parameters for the next training round and sending them to the interactive operation module 110. Based on this, the user can adjust their force application according to the guidance feedback during the current action, and in the next training round, use training target parameters adapted to their performance in this training session. Compared to devices that only record the number of presses or completion time, this embodiment can simultaneously reflect how the user completes the action and use the evaluation results for subsequent training.

[0034] Based on the above embodiments, and further considering the problem that single-point mechanical data is difficult to represent the finger contact position and pressure distribution changes, this embodiment provides a further description of the mechanical information acquisition unit 111. For example, the mechanical information acquisition unit 111 may include a pressure sensor array, which includes multiple pressure acquisition areas; the main control device is also used to obtain at least one spatial motion evaluation feature from the pressure distribution center, pressure application area, and pressure distribution center trajectory based on the pressure data corresponding to each pressure acquisition area, and to use the spatial motion evaluation feature as one of the multiple motion evaluation features.

[0035] Each pressure acquisition area corresponds to a spatial location, and the pressure data at that location is output at each sampling time.

[0036] The pressure sensor array can be a thin-film pressure sensor array, a resistive pressure sensor array, a capacitive pressure sensor array, or a piezoelectric pressure sensor array. Multiple pressure acquisition areas can form a rectangular grid, a circular grid, or an arrangement adapted to the shape of the pressure cap. Generally, before formal training, the zero-point values ​​of each pressure acquisition area can be obtained under no-load conditions. Then, the conversion relationship between each pressure acquisition area can be obtained using a known load to reduce output differences between different acquisition areas.

[0037] In one implementation, the main control device 120 can determine the pressure distribution center based on the pressure data of each pressure acquisition area at the same sampling time and the positional relationship of each pressure acquisition area, and determine the total force based on the pressure data of each pressure acquisition area. The pressure distribution center is used to represent the distribution location of pressure in multiple pressure acquisition areas, and the total force is used to represent the overall force applied by the user at that sampling time.

[0038] When the pressure data of a pressure acquisition area exceeds the area determination threshold, that pressure acquisition area can be included in the effective contact area. The pressure application area can be obtained based on the number of effective contact areas and the area of ​​a single pressure acquisition area. Alternatively, the pressure application area can also be obtained based on the outer contour formed by adjacent effective contact areas.

[0039] The main control device 120 can also connect the pressure distribution centers at each time point according to the sampling time sequence to obtain the trajectory of the pressure distribution center. The trajectory length can be obtained based on the distance between adjacent pressure distribution centers; the displacement per unit time can be obtained based on the relationship between the trajectory length and the continuous holding time; and the smoothness of the trajectory can be obtained based on the change in the direction of adjacent trajectories. At least one of the pressure distribution center, the pressure application area, and the trajectory of the pressure distribution center can be used as a spatial motion evaluation feature to determine whether the finger pressing position is stable, whether the contact area changes, or whether there is a significant offset.

[0040] For example, in training that requires the index fingertip to maintain pressure, if the total force meets the target requirement but the center of pressure distribution continues to move towards the edge of the button, the action might be judged as successful based solely on the total force. However, spatial motion evaluation features can reflect changes in the contact position. On the other hand, in multi-finger training, each interactive operation module 110 is equipped with a pressure sensor array, which can also obtain the pressure application area and force application sequence of different fingers. Thus, it can be seen that the pressure sensor array not only retains information on force changes over time but also adds spatial information.

[0041] In addition to the pressure distribution center, pressure application area, and pressure distribution center trajectory, the main control device 120 can also obtain the average force, force rise slope, force fall slope, steady-state error, compression rhythm, and the smoothness of the pressure distribution center trajectory, and select one or more of these as additional movement evaluation features according to the current rehabilitation goals. This embodiment does not imply that all features need to be used in every training session.

[0042] Building upon the above embodiments, this embodiment further addresses the issue that providing a result only after the action is completed cannot help the user correct their current force application. It explains the guidance feedback during the operation process. The target parameters include the target force and the allowable force range, which is defined by a lower limit and an upper limit. The processing unit obtains the current force based on the currently collected mechanical data. When the current force is less than the lower limit, it determines an insufficient force state; when the current force is greater than or equal to the lower limit and less than or equal to the upper limit, it determines a satisfactory force state; and when the current force is greater than the upper limit, it determines an excessive force state. The unit then controls the feedback unit to output differentiated guidance feedback for each of the insufficient, satisfactory, and excessive force states.

[0043] Processing unit 112 obtains the current force based on the currently collected mechanical data and compares the current force with the lower limit and upper limit of the force, respectively. When the current force is less than the lower limit, processing unit 112 determines that the force is insufficient; when the current force is greater than or equal to the lower limit and less than or equal to the upper limit, processing unit 112 determines that the force is adequate; when the current force is greater than the upper limit, processing unit 112 determines that the force is excessive. Feedback unit 113 provides guidance feedback that differentiates between the three states, allowing the user to adjust the force applied to their fingers before the action is completed.

[0044] For example, the feedback unit 113 can use a multi-colored light ring surrounding the pressing cap. Insufficient pressure corresponds to the first light state, adequate pressure to the second light state, and excessive pressure to the third light state. The first light state can be a solid red, the second light state can be a solid green, and the third light state can be a flashing blue. Color is optional and does not mean that the feedback unit 113 must use these colors. The feedback unit 113 can also use brightness changes, light flow changes, buzzing sounds, voice prompts, or vibration prompts, and can output two or more types of feedback simultaneously.

[0045] The processing unit 112 can perform analog-to-digital conversion and filtering on the raw signal from the pressure sensor array to obtain the current force. The filtering can employ moving average filtering, median filtering, or low-pass filtering. Furthermore, the processing unit 112 can remove data segments with durations shorter than the accidental touch time threshold, or remove data points with changes exceeding the abnormal change threshold, to reduce the impact of abnormal data caused by accidental touches or spasms on the judgment of the current operating status.

[0046] Since the judgment of the current operating state and the control of the feedback unit 113 are completed within the interactive operation module 110, the feedback unit 113 can still output guiding feedback based on the current force even when the main control device 120 has not yet received complete data. In one embodiment, the time from obtaining sampling data from the force information acquisition unit 111 to the feedback unit 113 changing its output state is less than 150 milliseconds. In addition to this method, the main control device 120 can also simultaneously display the current force curve, the target force band, and the hold countdown on the touch screen, so that the local feedback on the interactive operation module 110 corresponds to the display information on the main control device 120.

[0047] Furthermore, considering that a single maximum force cannot simultaneously represent force exertion capability, stability, and holding capability, this embodiment describes the formation process of the first action evaluation feature, the second action evaluation feature, and the third action evaluation feature. The operation target parameters include target force, allowable force range, and target holding time. The main control device is used to obtain the first action evaluation feature based on the relationship between the peak force during a single active operation and the target force. The main control device is also used to determine the first moment when the mechanical data first enters the allowable force range and the second moment when the mechanical data first leaves the allowable force range after the first moment, and to define the period between the first moment and the second moment as a continuous holding period. The second action evaluation feature is obtained based on the relationship between the mean force and the standard deviation of the force of the mechanical data within the continuous holding period, and the second action evaluation feature increases as the standard deviation of the force decreases relative to the mean force. The main control device is also used to obtain the continuous holding time based on the first moment and the second moment, and to obtain the third action evaluation feature based on the relationship between the continuous holding time and the target holding time.

[0048] like Figure 2 As shown, a single active operation can generate mechanical data showing the change of force over time. The intensity values ​​at each sampling point are as follows: , to ,Right now .

[0049] The main control device 120 can first process the mechanical data. Filtering and outlier processing are performed to obtain valid mechanical data. Peak force during a single active operation. The maximum force value among all sampling points represents the maximum active force exerted by the user in that action. The corresponding relationship can be written as: ; Target strength is In one implementation, the first motion evaluation feature is the peak force achievement rate. Furthermore, an upper limit is set for cases where the peak force exceeds the target force, preventing excessive force from further increasing the result. The corresponding relationship can be written as: ; In other words, when the peak force is lower than the target force, the peak force compliance rate increases with the peak force; when the peak force reaches or exceeds the target force, the peak force compliance rate is set to 1. This processing is suitable for evaluating whether the active force requirement has been met. If a training session also needs to evaluate the degree of excessive force, an overshoot characteristic can be obtained separately based on the difference between the peak force and the upper limit of force, or it can be formed by subtracting the tolerance from the target force to the target force plus the tolerance from the allowable range of force applied, where the tolerance is expressed as... The main control device 120 determines the first moment when the mechanical data enters the permissible force range. And determine the mechanical data at the first moment. The second moment after the first departure from the allowed range of force The first moment By the second moment The time interval between these intervals is considered a continuous holding period. This definition means that the continuous holding time ends when the user leaves the permissible force range, and the time spent re-entering the permissible force range afterward is not counted in the continuous holding period. .

[0050] During the continuous holding period, it has When sampling at each point, the main control device 120 first obtains the average intensity over the continuous holding period. Then obtain the standard deviation of the strength. The correspondence can be written as: ; ; In one implementation, the second motion evaluation feature adopts the force stability index. The smaller the standard deviation of force relative to the mean of force, the greater the force stability index. This relationship can be expressed as: ; When the force stability index is lower than the preset stability threshold, it can be determined that the force fluctuation during the holding phase is large. Alternatively, the coefficient of variation can be directly used... The evaluation value can be converted into a positive evaluation value, or a second action evaluation feature can be formed based on the maximum deviation during the continuous holding period. This embodiment does not limit the use of only one mathematical expression.

[0051] Continuous duration From the second moment With the first moment To obtain, that is: ; The target duration is In one implementation, the third action evaluation feature is target retention. And an upper limit is set after the continuous holding time reaches the target holding time. The corresponding relationship can be written as: ; The first action evaluation feature reflects whether the active force exertion reaches the target; the second action evaluation feature reflects the force fluctuation during the holding phase; and the third action evaluation feature reflects whether the user can continuously maintain the target force. Based on these three aspects, different situations can be distinguished, such as insufficient force exertion, reaching the target but maintaining it unstablely, and stable force but insufficient duration.

[0052] Based on the above embodiments, and further considering the issue that the first action evaluation feature, the second action evaluation feature, and the third action evaluation feature may adopt different numerical ranges, this embodiment explains the formation of the training evaluation result and the determination of the action evaluation feature to be improved. The main control device is used to normalize the first action evaluation feature, the second action evaluation feature, and the third action evaluation feature respectively, and then weights the normalized first action evaluation feature, the second action evaluation feature, and the third action evaluation feature according to preset weights to obtain the training evaluation result. The main control device is also used to obtain the action evaluation feature to be improved based on the relationship between the normalized first action evaluation feature, the second action evaluation feature, and the third action evaluation feature and their respective corresponding evaluation thresholds.

[0053] When the first, second, and third motion evaluation features are already within the range of 0 to 1, they can be directly used as the result after normalization. If one of the motion evaluation features is expressed as a percentage, the percentage value can be converted into a proportional value; if one of the motion evaluation features uses Newtons or seconds as the unit, it can be converted into a dimensionless value according to a preset reference range. The use of normalization in this embodiment does not mean that the preset range after normalization must be between 0 and 1.

[0054] The main control device 120 performs weighted processing on the three normalized action evaluation features according to preset weights to obtain the training evaluation result CRS. In one implementation, the correspondence can be written as: ; ; in, , and These are the preset weights for the first, second, and third movement evaluation features, respectively. In one optional mode, the sum of the three preset weights is 1. The weights can be input by the therapist or retrieved from a preset table based on the training stage. In the early stages of rehabilitation, when active exertion is the primary focus, this can be used... 0.6 0.2 The value is 0.2; during the mid-stage of rehabilitation, when the focus is on controlling stability, the following methods can be used. 0.2 0.6 The value is 0.2; in the late stage of rehabilitation, when the focus is on maintaining ability, the following methods can be used. 0.2 0.2 The value is 0.6. The above values ​​are a set of optional parameters and do not restrict other weight combinations.

[0055] Training evaluation results can be saved as a ratio value from 0 to 1, or converted to a score from 0 to 100 when displayed. A higher value indicates that the training performance is closer to the target. On the other hand, the main control device 120 retains three sub-item results and compares each sub-item result with its corresponding evaluation threshold. When a sub-item result is lower than its corresponding evaluation threshold, the corresponding action evaluation feature can be identified as an action evaluation feature that needs improvement.

[0056] When two or more sub-item results are all below their corresponding evaluation thresholds, the sub-item with the lowest value can be identified as the current action evaluation feature to be improved. Alternatively, selection can be based on the priority order set by the therapist. The main control device 120 can also display a radar chart formed by the first, second, and third action evaluation features on the touchscreen, highlighting the directions with lower values. For example, when the force stability index is below the evaluation threshold, a prompt to reduce force jitter can be displayed. Thus, the overall score is used to represent the overall performance, while the sub-item results are used to identify abilities that need further training.

[0057] Based on the above embodiments, further considering the potential impact of random conditions on a single training session, this embodiment generates a stage training evaluation result based on the training evaluation results corresponding to multiple consecutive training rounds. The main control device is used to acquire the training evaluation results corresponding to multiple consecutive training rounds, and obtains the stage training evaluation result based on the average value of the training evaluation results corresponding to multiple consecutive training rounds; when the stage training evaluation result is greater than or equal to the upper limit evaluation threshold, the main control device increases the target intensity, reduces the allowable range of intensity, or extends the target holding time; when the stage training evaluation result is greater than or equal to the lower limit evaluation threshold and less than the upper limit evaluation threshold, the main control device keeps the target intensity, the allowable range of intensity, and the target holding time unchanged.

[0058] The main control device 120 can save the training evaluation results corresponding to each training round and obtain the training evaluation result sequence from the most recent N training rounds. N can be 3 to 5, or it can be set by the therapist according to the training frequency.

[0059] In one implementation, the phase training evaluation results Use the most recent The average value of each training round can be expressed as follows: ; in, Indicates the first The training evaluation results for each training round. Using the average of multiple training rounds can reduce the impact of a single accidental touch, a brief lapse in attention, or a single abnormal exertion on parameter adjustments. Alternatively, a weighted average can be used for the training evaluation results at different stages, giving greater weight to more recent evaluation results.

[0060] The main control device 120 compares the phase training evaluation results with the upper and lower evaluation thresholds. If the training evaluation results are set to a score of 0 to 100, the upper evaluation threshold can be set to 85 and the lower evaluation threshold can be set to 60. If the training evaluation results are set to a ratio of 0 to 1, the corresponding thresholds can be set to 0.85 and 0.60.

[0061] When the evaluation result of a training phase is greater than or equal to the upper limit evaluation threshold, it indicates that the user has successfully completed the current training requirements. The main control device 120 can increase the target intensity by 5% to 10%, reduce the allowable intensity range by 10%, or extend the target holding time. A single parameter update can change only one parameter, or it can select parameters based on the evaluation characteristics of the action to be improved. For example, if the first action evaluation characteristic is high while the second action evaluation characteristic is close to the evaluation threshold, the allowable intensity range can be reduced first, without increasing the target intensity.

[0062] When the stage training evaluation result is between the lower and upper evaluation thresholds, the main control device 120 can maintain the target intensity, the allowable intensity range, and the target holding time unchanged, allowing the user to continue training under the current parameters. When the stage training evaluation result is below the lower evaluation threshold, the main control device 120 can reduce the target intensity by 5% to 10%, expand the allowable intensity range by 10%, or shorten the target holding time. This reduction processing can be used in the method embodiment or as another implementation of the system.

[0063] In addition, the active hand function rehabilitation training system can also be equipped with an auxiliary assistance unit connected to the main control device 120. When the training evaluation results are continuously below the lower limit evaluation threshold, the auxiliary assistance unit can provide a small amount of assistance force through finger supports or traction devices. The auxiliary assistance unit is an optional component; even without it, the system can still reduce training requirements by lowering the target force or expanding the allowable force range.

[0064] Furthermore, considering the problem that abstract evaluation values ​​are not easily understood by users, this embodiment maps action evaluation features to interactive tasks and task control parameters. The main control device 120 stores the correspondence between action evaluation features, interactive tasks, and task control parameters. The main control device is used to determine the target interactive task and the task control parameters of the target interactive task based on the action evaluation features to be improved and the correspondence, update the task control parameters of the target interactive task based on the mechanical data, and control the operation of the target interactive task according to the updated task control parameters; the main control device is also used to adjust the task control parameters for the next training round or replace the target interactive task based on the training evaluation results.

[0065] For example, when the first action evaluation feature is an action evaluation feature that needs improvement, a task that requires reaching the target force can be selected; when the second action evaluation feature is an action evaluation feature that needs improvement, a task that requires maintaining a stable output can be selected; when the third action evaluation feature is an action evaluation feature that needs improvement, a task that requires maintaining it for a period of time can be selected.

[0066] While processing the current task, the main control device 120 updates the task control parameters of the target interactive task based on the mechanical data, and controls the target interactive task to run according to the updated task control parameters. After training, the main control device 120 adjusts the task control parameters for the next training round based on the training evaluation results.

[0067] However, besides this method, a many-to-many relationship can also be used between motion evaluation features and interactive tasks. An interactive task can train both force exertion and stability simultaneously. The selection of interactive tasks can be based on user age, training stage, and operating habits. Interactive tasks are used to transform mechanical data into observable task variations without altering the evaluation relationship between mechanical data, motion evaluation features, and training target parameters.

[0068] like Figure 4 As shown, this embodiment provides an active hand function rehabilitation training method. This method can be executed by the aforementioned active hand function rehabilitation training system, or by other devices with corresponding data acquisition, feedback output, and data processing capabilities. The method can operate in a hospital, community, or home environment. The user completes active pressing, holding, and releasing actions through the interactive operation module 110. The method includes: S101: Obtain the current training target parameters and send the current training target parameters to the interactive operation module; the current training target parameters include operation target parameters and target evaluation features corresponding to multiple action evaluation features respectively; S102: Acquire mechanical data generated when the user actively operates the interactive operation module; S103: The interactive operation module obtains the current operation state based on the relationship between the currently collected mechanical data and the operation target parameters, and outputs corresponding guidance feedback based on the current operation state; S104: Based on the mechanical data, the plurality of motion evaluation features are obtained, wherein the plurality of motion evaluation features include at least a first motion evaluation feature characterizing active force exertion capability, a second motion evaluation feature characterizing force control stability, and a third motion evaluation feature characterizing continuous control capability. S105: Obtain training evaluation results based on the relationship between the plurality of action evaluation features and the corresponding target evaluation features; S106: Update the current training target parameters according to the training evaluation results, obtain the training target parameters corresponding to the next training round, and send the training target parameters corresponding to the next training round to the interactive operation module.

[0069] In step S101, the current training target parameters are obtained and sent to the interactive operation module 110. The current training target parameters include operation target parameters and target evaluation features corresponding to multiple action evaluation features. The current training target parameters can be derived from input by the therapist, a pre-saved training plan, or parameters updated in the previous training round.

[0070] In step S102, mechanical data generated when the user actively operates the interactive operation module 110 is acquired. The mechanical data can be force change data over time generated by a single sensor, or pressure data generated by multiple pressure acquisition areas in the pressure sensor array. During the acquisition process, the sampling time, the interactive operation module identifier, and the current training round identifier can be recorded simultaneously.

[0071] In step S103, the interactive operation module 110 determines the current operation state based on the relationship between the currently collected mechanical data and the operation target parameters, and outputs corresponding guidance feedback based on the current operation state. This step continues during user operation, allowing the user to adjust the current force according to the guidance feedback.

[0072] In step S104, multiple motion evaluation features are obtained based on the mechanical data. These multiple motion evaluation features include at least a first motion evaluation feature, a second motion evaluation feature, and a third motion evaluation feature, which are used to characterize active force generation ability, force control stability, and sustained control ability, respectively. Depending on training needs, at least one additional motion evaluation feature can also be obtained from the following: pressure distribution center, pressure application area, pressure distribution center trajectory, force rise slope, and operational rhythm.

[0073] In step S105, the training evaluation result is obtained based on the relationship between multiple action evaluation features and their corresponding target evaluation features. The training evaluation result can be obtained after a single action or after a set of actions. If a training set includes 10 presses, the sub-results for each press can be obtained first, and then the training evaluation result for this set can be obtained based on the sub-results of the 10 presses.

[0074] In step S106, the current training objective parameters are updated based on the training evaluation results to obtain the training objective parameters for the next training round, and these parameters are then sent to the interactive operation module 110. Afterward, the user performs a new active operation according to the training objective parameters for the next training round. In the above steps, the guidance feedback in step S103 is used for the current action, and the parameter update in step S106 is used for subsequent training rounds; both operate within different timeframes.

[0075] This embodiment further illustrates the process of obtaining multiple motion evaluation features and training evaluation results. The current training target parameters include target force, allowable force range, and target holding time. First, the peak force is determined from the mechanical data of a single active operation, and the first motion evaluation feature is obtained based on the relationship between the peak force and the target force. In one implementation, the peak force attainment rate described in the above embodiment can be used.

[0076] Subsequently, the first moment when the mechanical data first enters the permissible force range, and the second moment when the mechanical data first leaves the permissible force range after the first moment, are determined. The period between the first moment and the second moment is designated as the continuous holding period. If the mechanical data re-enters the permissible force range after the second moment, the time period formed by this re-entry is not included in the current continuous holding period and can be recorded separately as the next holding data segment.

[0077] Next, the second motion evaluation feature is obtained based on the relationship between the mean force and the standard deviation of force during the continuous holding period. The smaller the standard deviation of force relative to the mean force, the larger the second motion evaluation feature. For data with different sampling frequencies, the sampling points in the continuous holding period can be converted to the same sampling interval, or the time-weighted mean and time-weighted standard deviation can be obtained directly from the actual time of each sampling point.

[0078] The continuous holding time is then obtained based on the first and second moments, and a third motion evaluation feature is derived based on the relationship between the continuous holding time and the target holding time. If the first moment does not occur, it indicates that the motion did not enter the allowable force range, and the continuous holding time can be recorded as 0. If the first moment has occurred but the second moment has not occurred before the end of sampling, the end of sampling can be used as the second moment for this processing, or the sampling can be extended until the mechanical data leaves the allowable force range.

[0079] The first, second, and third action evaluation features are normalized respectively, and then weighted according to preset weights to obtain the training evaluation results. For the peak force attainment rate, force stability index, and target retention degree, which are already represented by 0 to 1, the numerical conversion can be omitted, and only the format uniformity and outlier detection are retained.

[0080] In addition, such as Figure 2 and Figure 3 As shown, the training evaluation results can also be associated with the displayed information. The main control device 120 can display the real-time force curve, peak force marker, target force band, continuous holding period, and holding countdown on the touch screen. After a single action is completed, a radar chart formed by the three action evaluation features and the training evaluation results can be displayed. When a certain action evaluation feature is lower than the corresponding evaluation threshold, the direction can be highlighted and a corresponding prompt can be displayed; when each action evaluation feature reaches the target, animation or voice information can be played.

[0081] This embodiment further illustrates the process of updating the current training objective parameters based on the training evaluation results. The main control device 120 acquires the training evaluation results corresponding to multiple consecutive training rounds, and obtains the stage training evaluation results based on the training evaluation results corresponding to multiple consecutive training rounds. The stage training evaluation results can be the average, weighted average, or median. When using the average, the most recent... The results of this training evaluation.

[0082] When the stage training evaluation result is greater than or equal to the upper evaluation threshold, increase the target intensity, narrow the allowable intensity range, or extend the target holding time. When the stage training evaluation result is less than the lower evaluation threshold, decrease the target intensity, expand the allowable intensity range, or shorten the target holding time. When the stage training evaluation result is between the two thresholds, the current parameters can be maintained. The parameter adjustment range can be set according to a fixed ratio or obtained based on the difference between the stage training evaluation result and the threshold.

[0083] The current training target parameters are updated based on the adjusted target intensity, allowable intensity range, or target duration to obtain the training target parameters for the next training round. The main control device 120 sends these parameters to the interactive operation module 110 and the target interactive task. The interactive operation module 110 updates the judgment criteria for the current operation state accordingly, and the target interactive task updates its task control parameters accordingly.

[0084] Furthermore, the main control device 120 can record the time of each parameter adjustment, the parameters before adjustment, the parameters after adjustment, the corresponding training evaluation results, and the evaluation characteristics of the movements to be improved, forming a training adjustment record. Therapists can check parameter changes and make manual modifications based on the training adjustment record. If the therapist inputs new training target parameters, the main control device 120 can use the new input parameters as the current training target parameters and start subsequent training from those parameters.

[0085] For the physical implementation of the interactive operation module 110, this embodiment provides an optional structure. A transparent pressure cap is provided on the top of the housing of the interactive operation module 110, a thin-film pressure sensor array is provided below the transparent pressure cap, a feedback light ring is provided around the pressure sensor array, a control board and a battery are provided inside the housing, and magnetic contacts are provided at the bottom of the housing. A microprocessor and a wireless communication unit are provided on the control board, and the microprocessor is connected to the pressure sensor array, the feedback light ring, and the wireless communication unit.

[0086] When the user presses the transparent pressure cap, the force is transmitted to the pressure sensor array. The transparent pressure cap can be made of a rigid transparent material or a light-transmitting material with a certain degree of elasticity. The light emitted by the feedback light ring can pass through the area around the transparent pressure cap or the side of the housing. The feedback light ring can use multi-color light-emitting diodes or multiple monochromatic light-emitting units.

[0087] The magnetic contacts can be a combination of magnets and spring contacts. The mounting plate has multiple interfaces compatible with the magnetic contacts. After the interactive operation module 110 is installed on any interface, it can obtain power through the contacts and perform wired communication. If the interactive operation module 110 has its own battery, it can also use only magnets for positioning and send mechanical data to the main control device 120 via a wireless communication unit.

[0088] Multiple interfaces can be arranged in a straight line, matrix, or arc. Therapists can install one or more interactive operation modules 110 in different positions depending on the finger length of the trainee and the training task. Installing one interactive operation module 110 allows for single-finger pressure training; installing multiple interactive operation modules 110 allows for multi-finger sequential operation, simultaneous operation, or two-handed operation. Thus, the interface positions of the mounting plate and the detachable structure of the interactive operation modules 110 allow the same device to adapt to different operating layouts.

[0089] However, besides this method, the interactive operation module 110 can also be fixedly mounted on the mounting plate, or connected via clips, connectors, or threaded parts. The mechanical information acquisition unit 111 can also be located in the grip ball, finger pinching component, or sliding operation component. As long as the mechanical data generated during the user's active operation can be acquired and guidance feedback can be output based on the current mechanical data, it can be used for the training described in this application.

[0090] For data processing and display, this embodiment provides an optional implementation method. After the pressure sensor array generates the raw signal, the processing unit 112 performs analog-to-digital conversion, zero-point correction, and filtering to obtain the current mechanical data, and immediately controls the feedback unit 113. The filtered current mechanical data and the raw array data can be sent to the main control device 120 simultaneously. When the main control device 120 saves the raw data, it can use other filtering parameters in subsequent processing to re-obtain the motion evaluation features.

[0091] The main control device 120 can divide a training session into a preparation phase, a force exertion phase, a continuous holding phase, and a release phase. The preparation phase is used to confirm the zero-load point; the force exertion phase is used to obtain the force rise slope and peak force; the continuous holding phase is used to obtain the mean force, standard deviation of force, and continuous holding duration; and the release phase is used to obtain the force fall slope. Additional features of different phases can be stored together with the first action evaluation features, second action evaluation features, and third action evaluation features.

[0092] The touchscreen can dynamically display the current force curve, target force baseline, allowable force range, peak force marker, continuous hold period, and hold countdown. After a single movement is completed, the touchscreen can display the first movement evaluation feature, the second movement evaluation feature, the third movement evaluation feature, and the training evaluation result. When the target is achieved, animation or audio information can be played; when the target is not achieved, the movement evaluation features that need improvement can be highlighted.

[0093] On the other hand, the main control device 120 can establish training records for each user, recording the date, training task, target intensity, allowable intensity range, target holding duration, movement evaluation characteristics, training evaluation results, and parameter adjustment results. These training records can be used by therapists to view changes between different dates and to determine the current training target parameters. The use of training records does not change the way current operational status is judged and feedback is handled within a single training session.

[0094] In summary, this application obtains mechanical data during the user's active operation through the interactive operation module 110, provides guidance feedback during the operation, and the main control device 120 generates training evaluation results from aspects such as force exertion ability, force control stability, and continuous control ability. These training evaluation results are then used to adjust subsequent training target parameters and interactive tasks. The sensor type, communication method, feedback form, parameter values, and interactive tasks in each embodiment can be selected according to the application environment; this embodiment does not impose strict limitations.

[0095] The active hand function rehabilitation training system and method provided in this application have been described in detail above. The various embodiments in the specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section. It should be noted that those skilled in the art can make several improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of this application.

[0096] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. An active hand function rehabilitation training system, characterized in that, include: At least one interactive operation module and a main control device; the interactive operation module includes a mechanical information acquisition unit, a processing unit and a feedback unit, the processing unit is connected to the mechanical information acquisition unit and the feedback unit respectively, and the main control device is communicatively connected to the interactive operation module; The main control device is used to acquire the current training target parameters and send the current training target parameters to the interactive operation module; the current training target parameters include operation target parameters and target evaluation features corresponding to multiple action evaluation features respectively; The mechanical information acquisition unit is used to collect mechanical data generated when the user actively operates the interactive operation module; The processing unit is used to obtain the current operation state based on the relationship between the currently collected mechanical data and the operation target parameters, and to control the feedback unit to output corresponding guidance feedback based on the current operation state; The main control device is used to acquire the mechanical data, obtain the plurality of action evaluation features based on the mechanical data, obtain training evaluation results based on the relationship between the plurality of action evaluation features and the corresponding target evaluation features, update the current training target parameters based on the training evaluation results, obtain the training target parameters corresponding to the next training round, and send the training target parameters corresponding to the next training round to the interactive operation module; wherein, the plurality of action evaluation features include at least a first action evaluation feature representing active force exertion ability, a second action evaluation feature representing force control stability, and a third action evaluation feature representing continuous control ability.

2. The active hand function rehabilitation training system according to claim 1, characterized in that, The mechanical information acquisition unit includes a pressure sensor array, which includes multiple pressure acquisition areas; The main control device is also used to obtain at least one spatial motion evaluation feature from the pressure distribution center, pressure action area and pressure distribution center trajectory based on the pressure data corresponding to each pressure acquisition area, and to use the spatial motion evaluation feature as one of the plurality of motion evaluation features.

3. The active hand function rehabilitation training system according to claim 1, characterized in that, The operational target parameters include the target force and the allowable force range, wherein the allowable force range includes a lower limit value and an upper limit value. The processing unit is used to obtain the current force based on the currently collected mechanical data, to obtain a force insufficient state when the current force is less than the force lower limit, to obtain a force qualified state when the current force is greater than or equal to the force lower limit and less than or equal to the force upper limit, and to obtain a force excessive state when the current force is greater than the force upper limit. The processing unit is also used to control the feedback unit to output guiding feedback that distinguishes between the force insufficient state, the force qualified state, and the force excessive state.

4. The active hand function rehabilitation training system according to claim 1, characterized in that, The operational target parameters include target force, allowable force range, and target holding time; The main control device is used to obtain the first action evaluation feature based on the relationship between the peak force and the target force during an active operation. The main control device is also used to determine the first moment when the mechanical data first enters the force allowable range and the second moment when the mechanical data first leaves the force allowable range after the first moment, and to take the time period between the first moment and the second moment as a continuous holding period; and to obtain the second action evaluation feature based on the relationship between the force mean and force standard deviation of the mechanical data in the continuous holding period, wherein the second action evaluation feature increases as the force standard deviation decreases relative to the force mean. The main control device is also used to obtain the continuous holding duration based on the first time and the second time, and to obtain the third action evaluation feature based on the relationship between the continuous holding duration and the target holding duration.

5. The active hand function rehabilitation training system according to claim 4, characterized in that, The main control device is used to normalize the first action evaluation feature, the second action evaluation feature and the third action evaluation feature respectively, and to perform weighted processing on the normalized first action evaluation feature, the second action evaluation feature and the third action evaluation feature according to preset weights to obtain the training evaluation result; The main control device is also used to obtain the action evaluation features to be improved based on the relationship between the normalized first action evaluation features, the second action evaluation features and the third action evaluation features and their respective corresponding evaluation thresholds.

6. The active hand function rehabilitation training system according to claim 5, characterized in that, The main control device is used to acquire the training evaluation results corresponding to multiple consecutive training rounds, and to obtain the stage training evaluation result based on the average value of the training evaluation results corresponding to multiple consecutive training rounds; when the stage training evaluation result is greater than or equal to the upper limit evaluation threshold, the main control device increases the target intensity, reduces the allowable range of the intensity, or extends the target holding time. When the training evaluation result of the stage is greater than or equal to the lower evaluation threshold and less than the upper evaluation threshold, the main control device keeps the target force, the allowable range of force, and the target holding time unchanged.

7. The active hand function rehabilitation training system according to claim 5, characterized in that, The main control device stores the correspondence between action evaluation features, interactive tasks, and task control parameters; The main control device is used to determine the target interaction task and the task control parameters of the target interaction task based on the evaluation features of the action to be improved and the correspondence, update the task control parameters of the target interaction task based on the mechanical data, and control the operation of the target interaction task according to the updated task control parameters. The main control device is also used to adjust the task control parameters for the next training round or change the target interactive task based on the training evaluation results.

8. A method for active hand function rehabilitation training, characterized in that, The method, applied to the active hand function rehabilitation training system according to any one of claims 1-7, comprises: Obtain the current training target parameters and send the current training target parameters to the interactive operation module; the current training target parameters include operation target parameters and target evaluation features corresponding to multiple action evaluation features respectively; Acquire mechanical data generated when the user actively operates the interactive module; The interactive operation module obtains the current operation state based on the relationship between the currently collected mechanical data and the operation target parameters, and outputs corresponding guidance feedback based on the current operation state; The plurality of motion evaluation features are obtained based on the mechanical data. The plurality of motion evaluation features include at least a first motion evaluation feature characterizing active force exertion capability, a second motion evaluation feature characterizing force control stability, and a third motion evaluation feature characterizing continuous control capability. The training evaluation result is obtained based on the relationship between the multiple action evaluation features and their corresponding target evaluation features; The current training objective parameters are updated based on the training evaluation results to obtain the training objective parameters for the next training round, and the training objective parameters for the next training round are sent to the interactive operation module.

9. The active hand function rehabilitation training method according to claim 8, characterized in that, The current training target parameters include target intensity, allowable intensity range, and target duration. The process of obtaining the multiple motion evaluation features based on the mechanical data includes: The first action evaluation feature is obtained based on the relationship between the peak force and the target force during an active operation. Determine the first moment when the mechanical data first enters the force allowable range and the second moment when the mechanical data first leaves the force allowable range after the first moment, and take the time period between the first moment and the second moment as the continuous holding period; The second motion evaluation feature is obtained based on the relationship between the mean force and the standard deviation of the force of the mechanical data during the continuous holding period. The second motion evaluation feature increases as the standard deviation of force decreases relative to the mean force. The continuous holding duration is obtained based on the first time point and the second time point, and the third action evaluation feature is obtained based on the relationship between the continuous holding duration and the target holding duration; The first action evaluation feature, the second action evaluation feature, and the third action evaluation feature are normalized respectively, and then the normalized first action evaluation feature, the second action evaluation feature, and the third action evaluation feature are weighted according to preset weights to obtain the training evaluation result.

10. The active hand function rehabilitation training method according to claim 8, characterized in that, Updating the current training target parameters based on the training evaluation results includes: Obtain the training evaluation results corresponding to multiple consecutive training rounds, and obtain the stage training evaluation results based on the training evaluation results corresponding to multiple consecutive training rounds. When the training evaluation result of the aforementioned stage is greater than or equal to the upper limit evaluation threshold, the target intensity is increased, the allowable range of the intensity is narrowed, or the target retention time is extended. When the training evaluation result of the aforementioned stage is less than the lower limit evaluation threshold, the target intensity is reduced, the allowable range of the intensity is expanded, or the target retention time is shortened. The current training target parameters are updated based on the adjusted target intensity, the allowable range of intensity, or the target duration, to obtain the training target parameters corresponding to the next training round.

Images