Remote orthopedic rehabilitation progress evaluation method and system based on digital twinborn

Abstract

The invention provides a remote orthopedic rehabilitation progress evaluation method and system based on digital twinning, and relates to the technical field of medical rehabilitation, and the method comprises the steps: obtaining the body data and rehabilitation exercise task data of a patient; inputting the body data and the rehabilitation exercise task data into a digital twin model; determining a state feature vector of the patient through the digital twin model; inputting the state feature vector into an orthopedic rehabilitation progress evaluation model; determining a rehabilitation progress score of the patient through the orthopedic rehabilitation progress evaluation model; determining a planning score difference degree according to the rehabilitation progress score, and dividing a rehabilitation exercise task according to the planning score difference degree; and adjusting the rehabilitation exercise task according to the division result. According to the invention, the accurate orthopedic rehabilitation progress can be provided, so that the orthopedic rehabilitation treatment is more efficient, flexible and accurate.

Description

Technical Field

[0001] This invention relates to the field of medical rehabilitation technology, and in particular to a remote orthopedic rehabilitation progress assessment method and system based on digital twins. Background Technology

[0002] The remote orthopedic rehabilitation progress assessment method based on digital twins uses digital twin technology to remotely monitor and assess the orthopedic rehabilitation process. This method combines the patient's body data with a virtual model to track their rehabilitation progress in real time, assess treatment effectiveness, provide personalized rehabilitation suggestions, and adjust the treatment plan according to the patient's rehabilitation status.

[0003] With the rapid development of telemedicine technology, digital twin technology can provide continuous and accurate rehabilitation monitoring by synchronizing the patient's actual physical condition with a virtual model in real time. This allows for timely reflection of the patient's rehabilitation progress and enables doctors to keep track of the patient's condition at any time, adjust treatment plans promptly, reduce the burden of patients traveling to and from the hospital, and improve treatment efficiency. This is of great significance for patients living in remote areas or with mobility difficulties.

[0004] However, traditional assessments rely on regular face-to-face examinations or patient self-reports, resulting in delayed information feedback and an inability to reflect the patient's recovery status in real time. The assessment process often depends on the doctor's subjective judgment, lacking objectivity and accuracy, and is prone to errors. In addition, existing methods lack the ability to monitor and adjust the dynamic changes during the recovery process in real time, and cannot be personalized according to the patient's actual needs, resulting in one-sided assessment results that cannot fully reflect the recovery progress. Summary of the Invention

[0005] In view of the shortcomings of the prior art, the purpose of this invention is to provide a remote orthopedic rehabilitation progress assessment method based on digital twins. This method can solve the problems of traditional assessment relying on regular face-to-face examinations or patient self-reports, which leads to delayed information feedback and an inability to reflect the patient's rehabilitation status in real time. The assessment process often relies on the doctor's subjective judgment, lacks objectivity and accuracy, and is prone to errors. In addition, existing methods lack the ability to monitor and adjust the dynamic changes during the rehabilitation process in real time, and cannot make personalized adjustments according to the patient's actual needs, resulting in one-sided assessment results that cannot fully reflect the rehabilitation progress.

[0006] A first aspect of this invention proposes a remote orthopedic rehabilitation progress assessment method based on digital twins, comprising:

[0007] S1: Acquire patient's physical data and rehabilitation exercise task data;

[0008] S2: Input the body data and the rehabilitation exercise task data into the digital twin model;

[0009] S3: Determine the patient's state feature vector using the digital twin model;

[0010] S4: Input the state feature vector into the orthopedic rehabilitation progress assessment model, wherein the orthopedic rehabilitation progress assessment model specifically includes: an autoencoder neural network module, a Gaussian mixture module, a fitting adjustment module, a performance measurement module, and a scoring mapping module;

[0011] S5: Determine the patient's rehabilitation progress score using the orthopedic rehabilitation progress assessment model;

[0012] S6: Based on the rehabilitation progress score, determine the planning score difference degree, and classify the rehabilitation exercise task into unqualified tasks, qualified tasks and excellent tasks according to the planning score difference degree.

[0013] S7: When the rehabilitation exercise task is an unqualified task, the optimal rehabilitation exercise task is re-formulated for the patient based on the digital twin model; when the rehabilitation exercise task is a qualified task, the rehabilitation exercise task is slightly adjusted; when the rehabilitation exercise task is an excellent task, the rehabilitation exercise task continues to be executed.

[0014] A second aspect of this invention provides a remote orthopedic rehabilitation progress assessment system based on digital twins, comprising: a processor and a memory;

[0015] The memory stores programs or instructions that can run on the processor, which, when executed by the processor, implement the steps of the remote orthopedic rehabilitation progress assessment method based on digital twins as described in the first aspect.

[0016] A third aspect of the present invention provides a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the remote orthopedic rehabilitation progress assessment method based on digital twins as described in the first aspect.

[0017] The beneficial effects of the technical solutions provided by the embodiments of the present invention include at least the following:

[0018] In this embodiment of the invention, the twin model enables continuous monitoring and precise assessment of the patient's rehabilitation process through digital twin modeling. This avoids the problem of delayed information feedback, allowing doctors to keep abreast of the patient's rehabilitation status and greatly improving the speed of medical response. The orthopedic rehabilitation progress assessment model adopts multiple modules and deeply analyzes the patient's rehabilitation progress based on state feature vectors, which can obtain accurate rehabilitation progress. This not only helps doctors identify specific problems in the patient's rehabilitation process, but also allows for timely adjustment of treatment plans, avoiding stagnation or mistreatment in the rehabilitation process. Based on the rehabilitation progress score, the patient's rehabilitation exercise tasks are dynamically adjusted, ensuring the personalization and precision of the rehabilitation plan, improving treatment effectiveness, and making orthopedic rehabilitation treatment more efficient, flexible, and precise. Attached Figure Description

[0019] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. Obviously, the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0020] Figure 1 This is a flowchart illustrating a remote orthopedic rehabilitation progress assessment method based on digital twins provided in an embodiment of the present invention.

[0021] Figure 2 This is a schematic diagram of a remote orthopedic rehabilitation progress assessment system based on digital twins provided in an embodiment of the present invention. Detailed Implementation

[0022] To enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. It should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0023] The following description, in conjunction with the accompanying drawings, details the remote orthopedic rehabilitation progress assessment method based on digital twins provided by the present invention through specific embodiments and application scenarios.

[0024] Reference manual attached Figure 1 The diagram illustrates a flowchart of a remote orthopedic rehabilitation progress assessment method based on digital twins provided by an embodiment of the present invention.

[0025] This invention provides a remote orthopedic rehabilitation progress assessment method based on digital twins, which may include the following steps:

[0026] S1: Acquire patient's physical data and rehabilitation exercise task data.

[0027] The patient's physical data refers to their physiological data, including motor function, joint angles, muscle strength, and neurological function, used to describe the patient's physical condition. Rehabilitation exercise task data refers to the specific data on various exercise tasks performed by the patient according to the rehabilitation treatment plan, including exercise type, intensity, number of repetitions, and duration, used to record the patient's rehabilitation training content and progress.

[0028] S2: Input body data and rehabilitation exercise task data into the digital twin model.

[0029] Digital twins refer to the real-time mapping and simulation of the state of physical entities or systems through virtual models. Digital twin models are virtual simulation models built based on actual patient data (such as body data and movement task data) to dynamically reflect changes in the patient's rehabilitation process.

[0030] It should be noted that by inputting the patient's physical data and rehabilitation exercise task data into the digital twin model, the patient's rehabilitation process can be accurately simulated using the model's real-time feedback. This allows for dynamic tracking of the patient's rehabilitation status through virtual simulation, avoiding the lagging assessment of the patient's condition in traditional methods. This enables doctors to understand the patient's actual rehabilitation progress in a timely manner and make precise adjustments to the treatment plan.

[0031] In one possible implementation, the digital twin model specifically includes: a musculoskeletal module, a neurological function impairment module, and a rehabilitation movement intention module.

[0032] S3: Determine the patient's state feature vector through a digital twin model.

[0033] Among them, the state feature vector is a multi-dimensional data vector generated by the digital twin model, which is used to describe the patient's current physical state and rehabilitation progress. It usually includes a quantitative representation of physiological parameters such as joint angles, muscle activation state, and motor ability, reflecting the patient's actual rehabilitation status.

[0034] It should be noted that the state feature vector generated by the digital twin model can accurately quantify and describe the patient's recovery status, transforming complex physiological states into quantifiable data. This allows doctors to understand the patient's recovery progress more clearly and intuitively, avoiding the errors that rely on doctors' subjective judgment in traditional methods.

[0035] In one possible implementation, the state feature vector includes: control torque generated on the musculoskeletal unit, total control torque generated on the joint, muscle activation state, predicted target angle of the joint, optimal muscle activation sequence, and the patient's self-awareness.

[0036] The control torque generated on a musculoskeletal unit refers to the torque (i.e., rotational force) generated in each musculoskeletal unit (such as the junction of a single muscle and bone). It is the rotational force exerted by the muscle on the bone, used to maintain or change the angle of the joint. The total control torque generated on the joint refers to the total rotational torque generated on the joint after the torques of each musculoskeletal unit are calculated through a mathematical model. It reflects the total force required for the entire joint movement and affects the joint's motion state. Muscle activation state refers to the degree of muscle activation, usually represented by measuring neural signals or physiological data. It reflects the intensity and functional state of muscle activity and directly affects motor performance and force output. The predicted joint target angle refers to the ideal angle that the joint should achieve, calculated based on rehabilitation goals or preset movement paths. This is a key parameter in the rehabilitation process, used to guide and evaluate the patient's movement effects. The optimal muscle activation sequence refers to the muscle activation order and intensity determined by an optimization algorithm under a given task and goal. The patient's self-awareness refers to the patient's self-perception of their physical condition, motor ability, pain sensation, etc.

[0037] In one possible implementation, S3 specifically includes:

[0038] S301: Determine the control torque and total control torque using the musculoskeletal module of the digital twin model.

[0039]

[0040]

[0041] in, This represents the torque generated on the i-th musculoskeletal unit, where j represents the muscle type number, j=+ indicates a synergistic muscle, and j=− indicates an antagonistic muscle. This represents the activation gain coefficient of the j-th muscle class in the i-th musculoskeletal unit. This represents the muscle length of the j-th type of muscle in the i-th musculoskeletal unit. This represents the muscle contraction velocity of the j-th type of muscle in the i-th musculoskeletal unit. This represents the activation state of the j-th muscle in the i-th musculoskeletal unit. This represents the passive gain coefficient of the j-th muscle class in the i-th musculoskeletal unit. This represents the total torque generated at joint p at time t. This represents the torque of the i-th musculoskeletal unit at time t. Let f represent the rotation matrix on the i-th musculoskeletal unit at time t, and let f represent the total number of musculoskeletal units.

[0042] A rotation matrix is ​​used to describe how torque is transferred from musculoskeletal units to joints, integrating local torque into global torque.

[0043] S302: Determine muscle activation status through the neurological function damage module.

[0044]

[0045] in, This represents the predicted activation level of the j-th muscle class in the i-th musculoskeletal unit. This indicates the degree of spasm in the j-th muscle within the i-th musculoskeletal unit. This represents the degree of weakness of the j-th type of muscle in the i-th musculoskeletal unit. This indicates the patient's cognitive bias regarding the degree of spasticity of the j-th type of muscle in the i-th musculoskeletal unit. This indicates the patient's cognitive bias regarding the degree of weakness of the j-th type of muscle in the i-th musculoskeletal unit.

[0046] The neurological impairment module illustrates the impact of neurological damage (such as spasticity and weakness) on muscle activation sequences, and incorporates patient perceptual biases. It primarily describes how to adjust the actual activation of muscles based on neurological damage and patient cognitive biases, thereby affecting motor performance. This indicates a patient's perception bias regarding their muscle spasms. Because patients' cognition is not always entirely accurate, they may underestimate or overestimate their spasms. This indicates a discrepancy in the patient's perception of their muscle weakness; the patient's perception of their weakness may not match their actual state of weakness. This refers to the muscle activation signals that a patient sets based on their motor intentions (such as wanting to perform a certain action) in the absence of nerve damage.

[0047] S303: Through the rehabilitation exercise intention module, determine the predicted target angle of the joint, the optimal activation sequence of the muscles, and the cognitive state.

[0048] In this embodiment of the invention, the core principle of the intent model is to plan and generate rehabilitation exercise activation sequences based on the patient's rehabilitation goals and historical rehabilitation exercise task performance. This module simulates how a patient generates muscle activation sequences based on their perceptual state, task feedback, and historical experience during the rehabilitation process to achieve specific rehabilitation goals.

[0049] In one possible implementation, the rehabilitation exercise intention module specifically includes: a perception unit, a planning unit, and a cognitive unit.

[0050] S303 specifically includes:

[0051] S3031: In the sensing unit, the predicted target angle of the joint is determined based on visual feedback and tactile feedback. Specifically, the visual feedback includes the target angle and current angle of the patient's joint, and the tactile feedback includes the torque of the end effector of the rehabilitation robot.

[0052]

[0053] in, Let β0 represent the predicted target angle of the patient's joint at time t, and ε represent the constant term. (t) N represents the prediction error at time t. vis Indicates the total length of a historical moment. The historical target angle at time t-kT represents the time interval of visual feedback, and β represents the time interval of visual feedback. k and α k Both represent the regression coefficients at the k-th historical moment, ε (t-kT) This represents the prediction error at time t-kT.

[0054] Specifically, by integrating visual and tactile feedback through the sensing unit, patients can be guided to complete rehabilitation exercises more accurately, improving the precision and flexibility of motor control and enhancing the patient's self-awareness. Visual feedback allows patients to see the difference between the target angle and the current angle in real time, helping them to better adjust their movements. Tactile feedback provides additional tactile guidance through the torque of the end effector of the rehabilitation robot, helping patients perceive the force requirements of the movement, thereby improving the effectiveness of rehabilitation training.

[0055] Patients receive visual feedback to obtain the target and current angles of their joints, thus understanding the gap between their current movement state and the target. Tactile feedback allows patients to receive torque from the robot's end effector, helping them adjust their movement.

[0056] S3032: In the planning unit, the optimal muscle activation sequence is determined based on the performance loss of the rehabilitation exercise task.

[0057]

[0058]

[0059]

[0060] in, This represents the optimal muscle activation sequence, and argmin represents finding the minimum value. This indicates a performance penalty for the task. This indicates the error in planning rehabilitation exercise tasks. Indicates muscle activation sequence, This represents the error in the planned rehabilitation exercise task at time t. This represents the predicted target angle of the patient's joint at time t. N represents the actual angle of the patient's joint at time t. fwd The value represents the size of the feed time window, and W represents the error weight matrix. Let R represent the muscle activation sequence at time t, and let R represent the activation cost matrix. T This indicates transpose.

[0061] Specifically, by using planning units to determine the optimal activation sequence of muscles based on the performance loss of rehabilitation exercise tasks, it is possible to ensure that muscles work together in the best way during rehabilitation training. By optimizing the activation sequence and intensity, the exercise effect can be maximized, the efficiency and accuracy of movements can be improved, ineffective movements can be reduced, and patients can achieve their rehabilitation goals faster and more efficiently.

[0062] S3033: In the cognitive unit, determine the cognitive state and the direction of cognitive state updates, wherein the cognitive state specifically includes the severity of spasticity and the severity of muscle weakness perceived by the patient.

[0063] The cognitive module is a core component of the motor intention model. Its main purpose is to adjust the patient's cognitive state by perceiving motor errors. Cognitive state refers to the patient's perception of their motor abilities and limitations, which is crucial for the rehabilitation process. The cognitive module helps patients gradually adapt to and adjust their motor strategies, especially when facing uncertainties in rehabilitation tasks. By providing real-time feedback and adjusting cognitive biases, it improves the patient's understanding of motor function, thereby promoting the recovery of motor function.

[0064] The cognitive module specifically includes:

[0065]

[0066]

[0067]

[0068] in, This represents the cognitive state at time t. This represents the cognitive state at time t-1. This represents the amount of cognitive update at time t. |A| represents the direction of cognitive update at time t, and |A| represents the magnitude of cognitive update. Indicates the performance loss associated with planned rehabilitation exercise tasks. Indicates the performance loss during actual rehabilitation exercise tasks. Let Softmax represent the Softmax activation function, and η represent the adjustment factor. The initial cognitive update direction is represented by N, the total duration of the rehabilitation exercise task is represented by Φ, and the probability distribution function is represented by Φ. ε represents the parameters of the Gaussian distribution at time t. t Let t represent the mean of the Gaussian distribution at time t, and Δ represent the increment.

[0069] Performance loss from planned rehabilitation exercise tasks It is measured by the error between the predicted target trajectory and the actual executed trajectory, and calculates the overall loss of the planned rehabilitation exercise task at time t. The performance loss of the actual rehabilitation exercise task is also considered. It is used to calculate the performance loss of actual rehabilitation exercise tasks. It is very similar to the calculation method of performance loss of planned rehabilitation exercise tasks. It calculates the error in the actual task execution process, that is, the error when performing the task.

[0070] S4: Input the state feature vector into the orthopedic rehabilitation progress assessment model, which specifically includes: an autoencoder neural network module, a Gaussian mixture module, a fitting adjustment module, a performance measurement module, and a scoring mapping module.

[0071] It should be noted that by inputting the state feature vector into the orthopedic rehabilitation progress assessment model, and through the comprehensive analysis of multiple modules within the model, the patient's rehabilitation progress can be assessed comprehensively and accurately. The autoencoder neural network module extracts key features through dimensionality reduction and data compression, the Gaussian mixture module handles complex multimodal data, the fitting and adjustment module optimizes the rehabilitation assessment model, the performance measurement module ensures the reliability of the assessment, and the scoring mapping module ultimately generates the rehabilitation score. The collaborative work of these multiple modules effectively reduces subjective errors, making the assessment results more scientific and accurate, and providing strong data support for subsequent adjustments to rehabilitation tasks and personalized treatment.

[0072] S5: Determine the patient's rehabilitation progress score using the orthopedic rehabilitation progress assessment model.

[0073] It should be noted that calculating a patient's rehabilitation progress score using the orthopedic rehabilitation progress assessment model can quantify the patient's rehabilitation status and provide objective and accurate progress indicators. Through scientific algorithms and data analysis, the model avoids subjective errors in traditional assessment methods, ensuring the consistency and reliability of the assessment results. The rehabilitation progress score can reflect the patient's rehabilitation progress in real time, helping doctors to fully understand the patient's recovery status, thereby making targeted treatment adjustments, optimizing rehabilitation plans, and improving treatment outcomes.

[0074] In one possible implementation, S5 specifically includes:

[0075] S501: In the autoencoder neural network module, the model parameters are reduced in dimensionality to generate low-dimensional features.

[0076] S502: Fitting low-dimensional features in the Gaussian mixture module.

[0077] The Gaussian mixture module is specifically as follows:

[0078]

[0079] in, Representing observation data The probability density given the parameter λ of the Gaussian mixture module. Let λ represent the observation data of the b-th rehabilitation exercise of patient a at time t, where λ = {π} c , 𝑐 ,Σ 𝑐}, where λ represents the set of all parameters in the Gaussian mixture module, and π c Let represent the mixing coefficient of the c-th Gaussian component, where C represents the total number of components in the Gaussian mixture module. Let represent the probability density function, 𝜇 𝑐 Let Σ represent the mean vector of the c-th Gaussian component. 𝑐 Let represent the covariance matrix of the c-th Gaussian component.

[0080] S503: In the fitting adjustment module, calculate the confidence level of the Gaussian mixture module. When the confidence level is greater than or equal to the preset confidence level, proceed directly to step S504. When the confidence level is less than the preset confidence level, increase the number of components in the Gaussian mixture module and return to step S502.

[0081] It should be noted that those skilled in the art can set the preset confidence level according to actual needs, and this invention does not limit this.

[0082] Specifically, by calculating the confidence level of the Gaussian mixture module and adjusting the fractions based on the confidence level, the accuracy and adaptability of the assessment model can be improved, and the complexity of the model can be adaptively adjusted. In addition, this dynamic adjustment mechanism ensures the high precision of the assessment results, enabling the model to flexibly respond to data changes of different patients or different rehabilitation stages, and further improve the reliability and accuracy of rehabilitation assessment.

[0083] The confidence level is as follows:

[0084]

[0085]

[0086] Where MC represents the confidence level and exp represents the exponential function. This represents the deviation of the motion pattern at time t, and max represents the maximum value taken. This represents the data from the b-th rehabilitation exercise of patient a at time t. Indicates the given parameter 𝜇 𝑐 and Σ 𝑐 At that time, data points The probability of occurrence.

[0087] S504: In the performance metrics module, measure the fit between the low-dimensional features and the Gaussian mixture module, and determine the performance metric value.

[0088] Specifically, by quantitatively evaluating the fit, the reliability and effectiveness of the model are ensured. By measuring the matching degree between low-dimensional features and the Gaussian mixture model, potential biases can be identified and adjusted in a timely manner, thereby optimizing the model's predictive ability.

[0089] The performance measurement module is specifically the log-likelihood measurement module.

[0090] The log-likelihood measurement module specifically includes:

[0091]

[0092] in, This represents the patient's rehabilitation exercise data given the model parameters λ. The log-likelihood value of Y a,b This indicates that this is the data for the bth rehabilitation exercise of the a-th patient, N represents the total duration of the rehabilitation exercise task, and log represents the logarithmic function.

[0093] S505: In the scoring mapping module, the performance metrics are mapped to obtain the rehabilitation progress score, which includes the rehabilitation progress score of the planned rehabilitation exercises and the rehabilitation progress score of the actual rehabilitation exercises.

[0094]

[0095] in, This represents the rehabilitation progress score based on the planned rehabilitation exercises, where e represents the base of the natural logarithm. The score represents the progress of actual rehabilitation exercises. This represents the performance metrics of planned rehabilitation exercises. This represents the performance metric of actual rehabilitation exercises, where μ represents the mean of the performance metric. α1 represents the standard deviation of the performance metric, α2 represents the hyperparameters corresponding to the planned rehabilitation exercises, and α3 represents the hyperparameters corresponding to the actual rehabilitation exercises.

[0096] S6: Based on the rehabilitation progress score, determine the planning score difference, and classify the rehabilitation exercise tasks into unqualified tasks, qualified tasks, and excellent tasks according to the planning score difference.

[0097] It should be noted that by calculating the planning score difference and classifying rehabilitation exercise tasks into unsatisfactory, satisfactory, and excellent tasks, the patient's current task completion status can be objectively reflected. This helps doctors to identify problems in the rehabilitation process in a timely manner, adjust the treatment plan in real time according to the patient's rehabilitation progress, ensure the personalization and targeting of treatment, and avoid patients from stagnation due to task mismatch.

[0098] In one possible implementation, S6 specifically includes:

[0099] S601: Calculate the difference between the rehabilitation progress score of the planned rehabilitation exercise and the rehabilitation progress score of the actual rehabilitation exercise, and determine the difference in the planned score.

[0100] S602: When the difference in the planning score is less than or equal to the first preset difference, the rehabilitation exercise task is classified as the excellent task; when the difference in the planning score is greater than the first preset difference and less than or equal to the second preset difference, the rehabilitation exercise task is classified as the qualified task; when the difference in the planning score is greater than the second preset difference, the rehabilitation exercise task is classified as the unqualified task.

[0101] It should be noted that those skilled in the art can set the magnitude of the first preset difference degree and the second preset difference degree according to actual needs, and the present invention does not limit this.

[0102] S7: When the rehabilitation exercise task is deemed unqualified, a new optimal rehabilitation exercise task is formulated for the patient based on the digital twin model. When the rehabilitation exercise task is deemed qualified, minor adjustments are made to the task. When the rehabilitation exercise task is deemed excellent, the task continues to be performed.

[0103] In this embodiment of the invention, a flexible task adjustment mechanism enables the rehabilitation program to be optimized in real time according to the patient's actual progress, ensuring that the patient is always in a task that is most suitable for their rehabilitation level. Whether it is to make remedial measures when the task is unsatisfactory or to make fine adjustments when the task is satisfactory, the treatment effect can be maximized and rehabilitation stagnation caused by excessive or insufficient task burden can be avoided. In addition, this dynamic feedback mechanism ensures the efficiency of the rehabilitation process and the active participation of the patient.

[0104] In one possible implementation, the step S7 of re-designing the optimal rehabilitation exercise task for the patient based on the digital twin model specifically involves:

[0105] Based on the digital twin model, and with the goal of minimizing the distance between the patient's cognitive state and neurological injury state, as well as the deviation in the patient's cognitive update direction, the rehabilitation exercise task is redesigned for the patient:

[0106]

[0107]

[0108]

[0109] Among them, Task t The optimal rehabilitation exercise task at time t is represented. This indicates the distance between the patient's cognitive state and their neurological damage state. This represents the directional deviation between the direction of cognitive updating and the target of neurological damage; S represents the patient's neurological damage status. and Indicates the weighting coefficient. This indicates the magnitude of the distance between the patient's cognitive state and their neurological damage state. It represents the Euclidean distance between the patient's cognitive state and neurological damage state.

[0110] In this embodiment of the invention, by formulating optimal rehabilitation exercise tasks, the rehabilitation effect can be maximized, avoiding patients from performing unsuitable exercise tasks under cognitive bias. By precisely adjusting the task content, patients can carry out rehabilitation training in the most suitable state, thereby improving rehabilitation efficiency, reducing fatigue and discomfort, and enhancing patients' participation and confidence.

[0111] The beneficial effects of the technical solutions provided by the embodiments of the present invention include at least the following:

[0112] In this embodiment of the invention, the twin model enables continuous monitoring and precise assessment of the patient's rehabilitation process through digital twin modeling. This avoids the problem of delayed information feedback, allowing doctors to keep abreast of the patient's rehabilitation status and greatly improving the speed of medical response. The orthopedic rehabilitation progress assessment model adopts multiple modules and deeply analyzes the patient's rehabilitation progress based on state feature vectors, which can obtain accurate rehabilitation progress. This not only helps doctors identify specific problems in the patient's rehabilitation process, but also allows for timely adjustment of treatment plans, avoiding stagnation or mistreatment in the rehabilitation process. Based on the rehabilitation progress score, the patient's rehabilitation exercise tasks are dynamically adjusted, ensuring the personalization and precision of the rehabilitation plan, improving treatment effectiveness, and making orthopedic rehabilitation treatment more efficient, flexible, and precise.

[0113] The remote orthopedic rehabilitation progress assessment method based on digital twins provided in this application can be implemented by a remote orthopedic rehabilitation progress assessment device based on digital twins. This application uses the execution of the remote orthopedic rehabilitation progress assessment method based on digital twins by the remote orthopedic rehabilitation progress assessment device based on digital twins as an example to illustrate the remote orthopedic rehabilitation progress assessment device based on digital twins provided in this application.

[0114] Reference manual attached Figure 2 The diagram shows a schematic representation of a remote orthopedic rehabilitation progress assessment system based on digital twins, provided by an embodiment of the present invention.

[0115] This invention provides a remote orthopedic rehabilitation progress assessment system 20 based on digital twins, including: a processor 201 and a memory 202;

[0116] The memory 202 stores programs or instructions that can run on the processor 201. When the program or instructions are executed by the processor 201, they implement the steps of the above-described remote orthopedic rehabilitation progress assessment method based on digital twins and achieve the same technical effect. To avoid repetition, the present invention will not elaborate further.

[0117] It should be understood that the processor 201 in this embodiment of the invention may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0118] It should also be understood that the memory 202 in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus RAM (DR RAM).

[0119] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0120] It should be understood that, in various embodiments of the present invention, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0121] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0122] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0123] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0124] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0125] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0126] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0127] This invention provides a readable storage medium comprising: storing a program or instructions on the readable storage medium, wherein when the program or instructions are executed by a processor, the program or instructions implement the steps of the above-described remote orthopedic rehabilitation progress assessment method based on digital twin, and can achieve the same technical effect. To avoid repetition, this invention will not elaborate further.

[0128] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the protection scope of the present invention.

Claims

1. A remote orthopedic rehabilitation progress assessment method based on digital twins, characterized in that, include: S1: Acquire patient's physical data and rehabilitation exercise task data; S2: Input the body data and the rehabilitation exercise task data into the digital twin model; S3: Determine the patient's state feature vector using the digital twin model; S4: Input the state feature vector into the orthopedic rehabilitation progress assessment model, wherein the orthopedic rehabilitation progress assessment model specifically includes: an autoencoder neural network module, a Gaussian mixture module, a fitting adjustment module, a performance measurement module, and a scoring mapping module; S5: Determine the patient's rehabilitation progress score using the orthopedic rehabilitation progress assessment model; S6: Based on the rehabilitation progress score, determine the planning score difference degree, and classify the rehabilitation exercise task into unqualified tasks, qualified tasks and excellent tasks according to the planning score difference degree. S7: When the rehabilitation exercise task is an unqualified task, the rehabilitation exercise task is re-formulated for the patient based on the digital twin model; when the rehabilitation exercise task is a qualified task, the rehabilitation exercise task is slightly adjusted; when the rehabilitation exercise task is an excellent task, the rehabilitation exercise task continues to be executed.

2. The remote orthopedic rehabilitation progress assessment method based on digital twins according to claim 1, characterized in that, The digital twin model specifically includes: a musculoskeletal module, a neurological function damage module, and a rehabilitation exercise intention module.

3. The remote orthopedic rehabilitation progress assessment method based on digital twins according to claim 1, characterized in that, The state feature vector includes: control torque generated on the musculoskeletal unit, total control torque generated on the joint, muscle activation state, predicted target angle of the joint, optimal muscle activation sequence, and the patient's self-awareness.

4. The remote orthopedic rehabilitation progress assessment method based on digital twins according to claim 3, characterized in that, S3 specifically includes: S301: Determine the control torque and the total control torque using the musculoskeletal module of the digital twin model; S302: Determine the muscle activation state through the neural function damage module of the digital twin model; S303: The rehabilitation movement intention module of the digital twin model is used to determine the predicted target angle of the joint, the optimal activation sequence of the muscles, and the cognitive state.

5. The remote orthopedic rehabilitation progress assessment method based on digital twins according to claim 4, characterized in that, The rehabilitation exercise intention module specifically includes: a perception unit, a planning unit, and a cognitive unit; Specifically, S303 includes: S3031: In the sensing unit, the predicted target angle of the joint is determined based on visual feedback and tactile feedback, wherein the visual feedback is specifically the target angle and current angle of the patient's joint, and the tactile feedback is specifically the end effector torque of the rehabilitation robot; S3032: In the planning unit, the optimal muscle activation sequence is determined based on the performance loss of the rehabilitation exercise task; S3033: In the cognitive unit, the cognitive state is determined, wherein the cognitive state specifically includes the severity of spasticity and the severity of muscle weakness perceived by the patient.

6. The remote orthopedic rehabilitation progress assessment method based on digital twins according to claim 1, characterized in that, S5 specifically includes: S501: In the autoencoder neural network module, the state feature vector is reduced in dimensionality to generate low-dimensional features; S502: Fit the low-dimensional features in the Gaussian mixture module; S503: In the fitting adjustment module, calculate the confidence level of the Gaussian mixture module. When the confidence level is greater than or equal to the preset confidence level, proceed directly to step S504; when the confidence level is less than the preset confidence level, increase the number of components in the Gaussian mixture module and return to step S502. S504: In the performance measurement module, the degree of fit between the low-dimensional features and the Gaussian mixture module is measured to determine the performance measurement value; S505: In the scoring mapping module, the performance metric value is mapped to obtain the rehabilitation progress score, wherein the rehabilitation progress score includes the rehabilitation progress score of the planned rehabilitation exercise and the rehabilitation progress score of the actual rehabilitation exercise.

7. The remote orthopedic rehabilitation progress assessment method based on digital twin as described in claim 6, characterized in that, S6 specifically includes: S601: Calculate the difference between the rehabilitation progress score of the planned rehabilitation exercise and the rehabilitation progress score of the actual rehabilitation exercise, and determine the difference degree of the planned score; S602: When the difference in the planning score is less than or equal to the first preset difference, the rehabilitation exercise task is classified as the excellent task; when the difference in the planning score is greater than the first preset difference and less than or equal to the second preset difference, the rehabilitation exercise task is classified as the qualified task; when the difference in the planning score is greater than the second preset difference, the rehabilitation exercise task is classified as the unqualified task.

8. The remote orthopedic rehabilitation progress assessment method based on digital twins according to claim 1, characterized in that, Specifically, in S7, the optimal rehabilitation exercise task is redefined for the patient based on the digital twin model: Based on the digital twin model, the rehabilitation exercise task is redesigned for the patient with the goal of minimizing the distance between the patient's cognitive state and the state of neurological damage, as well as the deviation in the direction of the patient's cognitive update.

9. A remote orthopedic rehabilitation progress assessment system based on digital twins, characterized in that, include: Processor and memory; The memory stores programs or instructions that can run on the processor, which, when executed by the processor, implement the steps of the remote orthopedic rehabilitation progress assessment method based on digital twins as described in any one of claims 1 to 8.

10. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the remote orthopedic rehabilitation progress assessment method based on digital twins as described in any one of claims 1 to 8.

Images