Orthopedic surgery postoperative rehabilitation auxiliary monitoring system based on artificial intelligence

Abstract

The application relates to the technical field of rehabilitation auxiliary monitoring, in particular to a postoperative rehabilitation auxiliary monitoring system for orthopedic surgery based on artificial intelligence, which comprises the following steps: acquiring electrical signals of leg muscle groups driven by nerves of normal and abnormal legs under different training times; determining an initial abnormal performance degree according to the deviation of activation time and activation intensity of both sides; analyzing the compensation of different muscles to determine abnormal analysis indexes of each muscle; determining a functional performance change curve according to the time sequence distribution of the abnormal analysis indexes of each muscle under different training times; and performing rehabilitation auxiliary monitoring according to the functional change curves of all muscles. The application can monitor the electrical signal change of the muscle in real time, analyze the compensation characteristics, find the detail change, recovery trend and potential rehabilitation problem in the rehabilitation process in time, and obtain more accurate monitoring results.

Description

Technical Field

[0001] This invention relates to the field of rehabilitation auxiliary monitoring technology, specifically to an artificial intelligence-based postoperative rehabilitation auxiliary monitoring system for orthopedic surgery. Background Technology

[0002] Postoperative rehabilitation in orthopedic surgery, a process requiring long-term monitoring and management, has seen significant improvements thanks to advancements in AI technology, particularly the combination of smart hardware and big data analytics. In recent years, with the increasing incidence of orthopedic diseases, especially the growing demand for rehabilitation after fractures and joint replacement surgeries, the rehabilitation process for orthopedic patients—including physical therapy, pain management, and muscle function recovery—is aided by monitoring patients' rehabilitation progress to help doctors and patients better track the process and adjust treatment plans accordingly.

[0003] After orthopedic surgery, due to factors such as surgical trauma (inflammation, swelling, pain), changes in intra-articular structures (such as changes in proprioception after ligament reconstruction), and joint effusion, the brain "perceives" joint instability or damage and actively weakens muscle contraction to "protect" the joint. This reflexively inhibits the neural drive signals of muscles that cross the joint (such as the quadriceps femoris), resulting in disordered muscle linkage signals after surgery.

[0004] Traditional rehabilitation monitoring methods mostly involve periodic questioning by nurses or doctors to assess patients' daily rehabilitation training, motor performance, and daily activities. This lacks specific, systematic, and quantitative analysis, failing to fully grasp the patient's post-operative training and recovery progress. These crucial aspects of the rehabilitation process are often overlooked. Consequently, subtle changes, recovery trends, potential rehabilitation problems, and overtraining phenomena cannot be detected in a timely manner, leading to inaccurate rehabilitation monitoring and analysis. Summary of the Invention

[0005] To address the technical problem of inaccurate rehabilitation monitoring and analysis due to the inability to detect rehabilitation abnormalities in a timely manner in related technologies, this invention provides an artificial intelligence-based postoperative rehabilitation auxiliary monitoring system for orthopedic surgery. The specific technical solution adopted is as follows:

[0006] This invention proposes an artificial intelligence-based postoperative rehabilitation monitoring system for orthopedic surgery, the system comprising:

[0007] The acquisition module is used to acquire the electrical signals of the nerve drive of the leg muscle groups in the normal and abnormal legs under different training sessions, and to determine the activation time and activation intensity of the leg muscles at each position on each side from the electrical signals.

[0008] The performance analysis module is used to perform delay analysis based on the activation time of the electrical signals of the muscles on both sides of the legs to determine the time deviation coefficient, and to determine the intensity deviation coefficient based on the difference in activation intensity of the electrical signals of the muscles on both sides of the legs. Combining the time deviation coefficient and the intensity deviation coefficient, the initial abnormal performance of the abnormal side of the leg is determined.

[0009] The compensation analysis module is used to identify the adjacent muscles of each muscle at the activation time based on the activation time of different muscles on the abnormal side of the leg; to perform compensation analysis by combining the abnormal performance of the analyzed muscles to determine the degree of abnormal compensation disorder; and to determine the abnormal analysis index of each muscle by combining the initial abnormal performance and the degree of abnormal compensation disorder.

[0010] The monitoring module is used to determine the functional performance change curve based on the temporal distribution of abnormal analysis indicators of each muscle in different training sessions; and to perform artificial intelligence-assisted rehabilitation monitoring based on the functional change curves of all muscles.

[0011] Furthermore, the step of performing delay analysis based on the activation times of the electrical signals of the muscles on both legs to determine the time deviation coefficient includes:

[0012] The time interval between the activation time of the electrical signal of each muscle and the activation time of the first muscle is determined as the delay time of the corresponding muscle.

[0013] The absolute value of the difference in the time delay of the muscles at the same location between the abnormal and normal legs is taken as the time delay degree of the corresponding muscles.

[0014] The mean time delay of the muscles at each position on the abnormal side of the leg was calculated and normalized as the time deviation coefficient.

[0015] Furthermore, determining the intensity deviation coefficient based on the difference in activation intensity of the electrical signals of the muscles on both legs includes:

[0016] The absolute value of the difference in activation intensity of muscle electrical signals at the same location is used as the intensity difference of the corresponding muscle.

[0017] The mean value of the muscle strength difference at each position on the abnormal side of the leg was calculated and normalized as the strength deviation coefficient.

[0018] Furthermore, the determination of the initial abnormality level of the abnormal leg by combining the time deviation coefficient and the intensity deviation coefficient includes:

[0019] The mean of the time deviation coefficient and the intensity deviation coefficient is calculated as the initial abnormality level of the abnormal leg.

[0020] Furthermore, the step of determining the adjacent muscles of each muscle at the activation time based on the activation time of different muscles in the abnormal leg as the analyzed muscles includes:

[0021] The activation sequence is obtained by sorting all muscles in the abnormal leg according to the order of activation.

[0022] On the activation sequence, identify the two other muscles that are closest to any muscle sequence as the muscles to be analyzed.

[0023] Furthermore, the compensation analysis, which combines the analysis of abnormal muscle function manifestations to determine the degree of abnormal compensation disorder, includes:

[0024] Calculate the mean of abnormal functional performance of all analyzed muscles, and normalize the negative of the mean as the degree of abnormal compensation disorder.

[0025] Furthermore, the method of combining the initial abnormality level and the abnormal compensation disorder level to determine the abnormality analysis index for each muscle includes:

[0026] The abnormal compensation disorder is linearly mapped to the value range of [0.5, 1.5] to obtain the compensation weight;

[0027] The product of the initial abnormality level and the compensation weight is calculated as the anomaly analysis index.

[0028] Furthermore, determining the functional performance change curve based on the temporal distribution of abnormal analysis indicators for each muscle in different training sessions includes:

[0029] The abnormality analysis sequence is obtained by sorting the abnormality analysis indicators of each muscle in all training sessions.

[0030] A two-dimensional rectangular coordinate system is constructed with the number of occurrences as the x-axis and the anomaly analysis index as the y-axis to determine the coordinates of each element in the anomaly analysis sequence in the two-dimensional rectangular coordinate system.

[0031] Curve fitting was performed on all coordinate points using the least squares method to obtain the functional performance change curve.

[0032] Furthermore, the AI-assisted rehabilitation monitoring based on the functional change curves of all muscles includes:

[0033] Based on the fluctuations in the functional change curves of all muscles, target muscles for key abnormality monitoring were selected.

[0034] Calculate the mean of the abnormality analysis index for all target muscles in the current training session, and normalize the negative of the mean as the degree of recovery.

[0035] Furthermore, based on the fluctuations in the functional change curves of all muscles, the target muscles for focused abnormality monitoring are selected, including:

[0036] The functional change curves of all muscles are fitted with straight lines to obtain fitted straight lines, and the slope of the fitted straight lines is determined.

[0037] The average value of the ordinate of all coordinate points in the functional change curve is calculated to obtain the overall fluctuation coefficient.

[0038] The product of the overall fluctuation coefficient and the slope of any muscle is normalized and used as a monitoring indicator.

[0039] Muscles whose monitored indicators are greater than a preset threshold are designated as target muscles.

[0040] The present invention has the following beneficial effects:

[0041] This invention combines electrical signals from the normal and abnormal legs during training. It performs delay analysis based on the activation time of the electrical signals to determine a time deviation coefficient, and determines an intensity deviation coefficient based on the difference in activation intensity. Based on these two coefficients, the initial abnormality level of the abnormal leg is determined. Compared to related technologies that directly use interrogative recovery assessments and various discrete examination indicators, this invention combines continuous time-series analysis of data changes during exercise to construct an initial abnormality level that more closely approximates the true recovery state. Then, it analyzes the compensatory behavior of other muscles to determine the degree of abnormal compensation disorder. Combining the initial abnormality level and the degree of abnormal compensation disorder, it determines the abnormality analysis index for each muscle. This accurately captures the abnormal nature of postoperative muscle coordination, analyzes the compensatory effects of multiple muscle linkage combinations, and thus corrects the initial abnormality level, obtaining more accurate and realistic abnormality analysis indicators. Finally, it combines the changes in abnormality analysis indicators under different training sessions for rehabilitation-assisted monitoring. In summary, this invention can monitor changes in muscle electrical signals in real time and analyze them through compensatory characteristics, thereby promptly identifying subtle changes, recovery trends, and potential rehabilitation problems during the rehabilitation process and obtaining more accurate monitoring results. Attached Figure Description

[0042] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1A structural diagram of an artificial intelligence-based postoperative rehabilitation monitoring system for orthopedic surgery is provided in one embodiment of the present invention.

[0044] Figure 2 This is a schematic diagram of the functional performance change curve provided in one embodiment of the present invention. Detailed Implementation

[0045] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of an artificial intelligence-based postoperative rehabilitation monitoring system for orthopedic surgery proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0046] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0047] The following description, in conjunction with the accompanying drawings, details the specific solution of an artificial intelligence-based postoperative rehabilitation monitoring system for orthopedic surgery provided by this invention.

[0048] Please see Figure 1 The diagram illustrates a structural diagram of an artificial intelligence-based postoperative rehabilitation monitoring system for orthopedic surgery, according to an embodiment of the present invention. The system includes: an acquisition module 101, a performance analysis module 102, a compensation analysis module 103, and a monitoring module 104.

[0049] The acquisition module 101 is used to acquire the electrical signals of the nerve drive of the leg muscle groups in the normal and abnormal legs under different training sessions, and to determine the activation time and activation intensity of the leg muscles at each position on each side from the electrical signals.

[0050] Postoperative rehabilitation in orthopedic surgery, a process requiring long-term monitoring and management, has seen significant improvements thanks to the combination of smart hardware and big data analytics. In recent years, with the increasing incidence of orthopedic diseases, especially the growing demand for rehabilitation after fractures and joint replacement surgeries, the rehabilitation process for orthopedic patients, including physical therapy, pain management, and muscle function recovery, benefits from monitoring. This helps doctors and patients better track the rehabilitation progress and adjust treatment plans accordingly.

[0051] After orthopedic surgery, due to factors such as surgical trauma (inflammation, swelling, pain), changes in intra-articular structures (such as changes in proprioception after ligament reconstruction), and joint effusion, the brain "perceives" joint instability or damage and actively weakens muscle contraction to "protect" the joint. This reflexively inhibits the neural drive signals of muscles that cross the joint (such as the quadriceps femoris), resulting in disordered muscle linkage signals after surgery.

[0052] Traditional rehabilitation monitoring methods mostly involve periodic questioning by nurses or doctors to assess patients' daily rehabilitation training, motor performance, and daily activities. This lacks specific, systematic, and quantitative analysis, failing to fully grasp the patient's post-operative training and recovery progress. These crucial aspects of the rehabilitation process are often overlooked. Consequently, subtle changes, recovery trends, potential rehabilitation problems, and overtraining phenomena cannot be detected in a timely manner, leading to inaccurate rehabilitation monitoring and analysis.

[0053] Postoperative rehabilitation monitoring is crucial for orthopedic patients, as it determines how doctors assess and adjust training methods based on the patient's recovery progress. Furthermore, the specific rehabilitation training plan and monitoring focus will vary depending on the surgical site. Therefore, this invention focuses on the rehabilitation monitoring and analysis of patients who have undergone orthopedic surgery on one thigh (to specifically target the patient group aiming for high-quality recovery of leg movement after surgery).

[0054] Firstly, for patients, an intelligent wearable device (i.e. a lightweight wireless sEMG device) is used. By integrating electrode pads into a miniature wireless transmission module and attaching them directly to the skin (without excessively affecting the patient's normal training activities or causing any burden), the electrical signals of the patient's muscles are obtained.

[0055] A single, repetitive movement can be considered as one training session. For example, walking, which involves raising and lowering the leg, can be considered as one training session. The electrical signals of the neural drive of the leg muscle groups can be obtained from different training sessions. It should be noted that since the analysis focuses on both legs, the movement must be repetitive and repeated multiple times, and the movements must be identical on both legs. For example, walking at a steady pace, raising both legs while seated, and other common movements are not restricted in this regard.

[0056] The electrical signal of each muscle is at a minimum value when it is not activated, but a large electrical signal intensity is generated when it is activated. Therefore, the initial moment when the electrical signal intensity is greater than the preset intensity threshold is taken as the activation moment. At the same time, the average value of the electrical signal intensity within the duration range of the electrical signal intensity being greater than the preset intensity threshold is taken as the activation intensity.

[0057] The preset intensity threshold is a threshold value for determining the activation of an electrical signal. In this embodiment of the invention, the preset intensity threshold can be, for example, 200µV, and can be adjusted according to the actual situation without limitation.

[0058] The performance analysis module 102 is used to perform delay analysis based on the activation time of the electrical signals of the muscles on both sides of the legs, determine the time deviation coefficient, determine the intensity deviation coefficient based on the difference in activation intensity of the electrical signals of the muscles on both sides of the legs, and combine the time deviation coefficient and the intensity deviation coefficient to determine the initial abnormal performance of the abnormal side of the leg.

[0059] When the body performs a repetitive movement, under normal circumstances, the order, intensity, and duration of muscle activation in the left and right legs are relatively consistent. However, due to the effects of surgery, the muscles on the abnormal side may have lower strength and endurance than those on the normal side during rehabilitation, leading to delayed activation and intensity deviation. Therefore, delay analysis can be performed by analyzing the activation timing, and intensity deviation analysis can be performed by analyzing the difference in activation intensity of electrical signals, thereby determining the initial degree of abnormality in the abnormal leg compared to the normal leg.

[0060] Furthermore, in some embodiments of the present invention, a time deviation coefficient is determined by performing a delay analysis based on the activation time of the electrical signals of the muscles on both sides of the leg. This includes: determining the time interval between the activation time of the electrical signal of each muscle and the activation time of the first muscle, as the delay time of the corresponding muscle; taking the absolute value of the difference between the delay times of the muscles at the same position on the abnormal side leg and the normal side leg as the time delay degree of the corresponding muscle; calculating the mean of the time delay degree of each muscle at each position on the abnormal side leg, and normalizing it as the time deviation coefficient.

[0061] In this embodiment of the invention, since the two legs may be in different states of motion at the same time during a specific movement, for example, when walking, the left leg lands while the right leg is raised, the activation time of the first muscle of each leg is taken as the starting time of the corresponding side leg, and the time interval between the activation time of other muscles and the starting time is taken as the delay time of each muscle.

[0062] During normal walking, the activation sequence of muscles in both legs is the same. Taking walking as an example:

[0063] (1) Heel strike phase: Tibialis anterior (contracts to control the speed of foot descent) → Gluteus medius (stabilizes the pelvis) → Quadriceps femoris (especially the vastus medialis, which stabilizes the knee joint).

[0064] (2) Mid-standing phase: Gastrocnemius (controls tibial anterior movement and maintains forward body movement) → Gluteus maximus and hamstrings (extends hip and propels forward).

[0065] (3) Push-off phase: Triceps calf muscle (strong contraction to propel on tiptoe) → Quadriceps femoris muscle (stabilizes the knee in the final stage) → Iliopsoas muscle (initiates swing);

[0066] (4) Swing phase: Iliopsoas and rectus femoris (hip flexion) → Tibialis anterior (foot dorsiflexion to prevent dragging) → Hamstrings (deceleration and knee extension in the final stage to prepare for landing).

[0067] Since both legs have the same structure and the same muscles, by analyzing the delay time of the same muscles in both legs, the absolute value of the difference between the delay time of the abnormal leg and the normal leg in the same position of the muscle is taken as the time delay degree of the corresponding muscle. The larger the value of the time delay degree, the greater the difference in the activation time of the abnormal leg and the normal leg in the same position of the muscle. This situation may be due to abnormal muscle control caused by injury in the abnormal leg.

[0068] By integrating the time delays of muscles at all locations in the leg as a whole, the mean time delay of each muscle location on the abnormal side of the leg was calculated and normalized to serve as a time deviation coefficient. The time deviation coefficient can effectively characterize the activation time deviation of the abnormal side relative to the normal side. The larger the value, the more abnormal the muscle activation on the abnormal side itself, indicating that the muscle connections between muscles in the patient have not yet fully recovered to a normal state after surgery.

[0069] Furthermore, in some embodiments of the present invention, determining the intensity deviation coefficient based on the difference in activation intensity of electrical signals of muscles on both sides of the leg includes: taking the absolute value of the difference in activation intensity of muscle electrical signals at the same location as the intensity difference degree of the corresponding muscle; calculating the mean value of the intensity difference degree of each muscle at each location on the abnormal side of the leg, and normalizing it as the intensity deviation coefficient.

[0070] The analysis of the intensity deviation coefficient is similar to that of the time deviation coefficient. Intensity deviation is determined by the difference in activation intensity of muscle electrical signals at the same location. Then, the intensity deviations of muscles at all locations are integrated to obtain the intensity deviation coefficient. The deviation in activation intensity between the normal and abnormal sides of the muscles can serve as an indicator of neural response or functional problems of muscle cells. A larger intensity deviation coefficient indicates that the muscle is unable to provide the required electrical signal intensity in a short time or is providing excessive intensity, requiring greater attention. It reflects the abnormal side of the muscle's rapid fatigue or over-excitation, suggesting possible functional abnormalities.

[0071] In summary, by combining the time deviation coefficient and the intensity deviation coefficient, the initial abnormality level of the abnormal leg is determined, including: calculating the mean of the time deviation coefficient and the intensity deviation coefficient as the initial abnormality level of the abnormal leg.

[0072] Of course, in other embodiments of the present invention, different weight values ​​may be assigned to the time deviation coefficient and the intensity deviation coefficient respectively, and they may be weighted and summed to obtain the initial abnormal performance degree. For example, the weight of the time deviation coefficient is 0.4 and the weight of the intensity deviation coefficient is 0.6. There is no limitation on this.

[0073] It is understandable that the larger the value of the time deviation coefficient, the greater the difference between the muscle activation time and the normal situation; the larger the value of the intensity deviation coefficient, the greater the difference between the muscle activation intensity and the normal situation. Both indicate that the actual training performance of the abnormal leg is more abnormal. This characteristic is characterized by quantifying the initial abnormal performance.

[0074] The compensation analysis module 103 is used to determine the adjacent muscles of each muscle at the activation time based on the activation time of different muscles on the abnormal side of the leg; to perform compensation analysis by combining the abnormal performance of the analyzed muscles to determine the degree of abnormal compensation disorder; and to determine the abnormal analysis index of each muscle by combining the initial abnormal performance and the degree of abnormal compensation disorder.

[0075] During rehabilitation training, different exercises activate different muscle synergies, meaning each exercise corresponds to a different muscle response sequence. While these response sequences involve different muscles, in some cases, locally similar muscle synergies may occur, meaning the same muscle synergy may appear repeatedly in different exercises. However, the muscles preceding and following the same muscle synergy may differ under different muscle response sequences, thus leading to variations in functional performance for each exercise. Furthermore, due to the varying recovery rates of different muscles, compensatory mechanisms may emerge, resulting in functional differences in the same muscle synergy across different exercises.

[0076] Thus, the initial abnormality of each muscle on the abnormal side of the patient analyzed through the above process still has a certain deviation in representing the abnormal condition of each muscle. It is also necessary to consider the compensation between different muscles under different training movements in order to conduct a more accurate assessment.

[0077] Muscle compensation is a complex process that is related to the progress of muscle recovery. When the recovery of muscles is poor or when the neural connections between muscles have not been re-established, muscle compensation or being compensated for may occur.

[0078] For example, if the recovery of a preceding muscle synergy is poor, it will affect the synergy of subsequent muscles, causing abnormal performance in the current muscle synergy. Conversely, if the recovery of the current muscle synergy is poor, the response of the subsequent muscle synergy to stimulation will be affected, and it may also compensate for the current muscle synergy. Therefore, it is necessary to analyze the compensatory synergy of each muscle in temporal proximity.

[0079] First, it is necessary to identify the muscles affected by compensation. Further, in some embodiments of the present invention, based on the activation times of different muscles in the abnormal leg, other muscles adjacent to each muscle at the activation time are identified as the muscles to be analyzed. This includes: sorting all muscles in the abnormal leg according to the order of activation times to obtain an activation sequence; and identifying the two muscles closest to any muscle sequence in the activation sequence as the muscles to be analyzed.

[0080] In other words, in this embodiment of the invention, muscles with similar temporal activation are directly selected as the muscles to be analyzed for compensatory effects. In other embodiments of the invention, the information of the muscles to be analyzed for compensation can be determined based on actual physiological characteristics.

[0081] Then, based on the analysis of the abnormal conditions of the muscles, the degree of disorder of the abnormal scenario of each muscle is determined. This degree of disorder is specifically characterized as the degree of abnormal compensation disorder. Furthermore, in some embodiments of the present invention, compensation analysis is performed in conjunction with the analysis of the abnormal performance of muscle function to determine the degree of abnormal compensation disorder.

[0082] It is understandable that the lower the functional abnormality of the analyzed muscle (e.g., the previous activated muscle and the next activated muscle), the more it conforms to its normal state. Then the abnormality of the current muscle in the middle is closer to its true performance and has higher credibility. Therefore, in this embodiment of the invention, the mean of the abnormal functional performance of all analyzed muscles is calculated, and the negative of the mean is normalized as the degree of abnormal compensation disorder.

[0083] In this way, the degree of abnormal compensation disorder of each current muscle in the analyzed muscle group can be determined. The larger the value, the more complex the response relationship between the current muscle and the analyzed muscle group. This involves the differences in the recovery progress of different muscles and the complex situation of mutual compensation between different muscles.

[0084] Among them, the degree of abnormal compensation disorder indicates the degree of disorder. The higher the degree of abnormality of other muscles affected by compensation (the higher the degree of abnormal function performance), the more easily the current muscle is affected by the compensation of other abnormal muscles, that is, its credibility is low. In this case, more obvious abnormal performance is needed to prove that it is a truly abnormal muscle.

[0085] Therefore, in this embodiment of the invention, the degree of abnormal compensation disorder can be used as the weight of the initial abnormal performance degree to achieve weighted analysis.

[0086] Furthermore, in some embodiments of the present invention, the abnormality analysis index of each muscle is determined by combining the initial abnormality performance degree and the abnormal compensation disorder degree, including: linearly mapping the abnormal compensation disorder degree to the value range of [0.5, 1.5] to obtain the compensation weight; and calculating the product value of the initial abnormality performance degree and the compensation weight as the abnormality analysis index.

[0087] To facilitate weight analysis, the abnormal compensation disorder is first linearly mapped to a value range of [0.5, 1.5]. Of course, in other embodiments of the present invention, the mapping rules can be adjusted according to the actual situation, such as using an activation function for mapping, or using other mapping ranges, without limitation.

[0088] Then, the initial abnormal performance is weighted to obtain the abnormality analysis index. In this embodiment of the invention, the abnormality analysis index is the abnormal performance of the muscle function relative to the actual situation. The larger the value of the abnormality analysis index, the more abnormal the function of the muscle itself is from the perspective of compensation.

[0089] The monitoring module 104 is used to determine the functional performance change curve based on the temporal distribution of abnormal analysis indicators of each muscle in different training sessions; and to perform artificial intelligence rehabilitation-assisted monitoring based on the functional change curves of all muscles.

[0090] In this embodiment of the invention, the training sessions are repeated multiple times. Since the overall training difficulty is low, training can be conducted once every day. This allows for the statistical analysis of abnormal muscle indicators across all training sessions, and rehabilitation monitoring can be performed based on their temporal changes.

[0091] Furthermore, in some embodiments of the present invention, the functional performance change curve is determined based on the temporal distribution of the abnormal analysis indicators of each muscle in different training sessions, including: sorting the abnormal analysis indicators of each muscle in all training sessions to obtain an abnormal analysis sequence; constructing a two-dimensional rectangular coordinate system with the number of training sessions as the abscissa and the abnormal analysis indicators as the ordinate, and determining the coordinate point of each element in the abnormal analysis sequence in the two-dimensional rectangular coordinate system; and performing curve fitting on all coordinate points based on the least squares method to obtain the functional performance change curve.

[0092] See Figure 2 , Figure 2This is a schematic diagram of a functional performance change curve provided in one embodiment of the present invention. In this embodiment, the abnormality analysis indicators of all training sessions are sorted and analyzed over time, and curve fitting is performed to obtain the functional performance change curve. This functional performance change curve represents the abnormal changes of muscles under multiple training sessions. If the recovery effect is better, the functional performance change curve shows a more obvious downward trend, while if no recovery effect is produced, the functional performance change curve tends to be stable and unchanged.

[0093] Specifically, in some embodiments of the present invention, artificial intelligence-assisted rehabilitation monitoring is performed based on the functional change curves of all muscles, including: selecting target muscles to be monitored for key abnormalities based on the fluctuations of the functional change curves of all muscles; calculating the mean of the abnormality analysis index of all target muscles in the current training session, and normalizing the negative of the mean as the degree of rehabilitation.

[0094] First, based on the fluctuation information of the functional change curves, the target muscles for key abnormal monitoring are identified, including: fitting a straight line to the functional change curves of all muscles to obtain a fitted straight line and determining the slope of the fitted straight line; calculating the mean value of the ordinate of all coordinate points in the functional change curves to obtain the overall fluctuation coefficient; normalizing the product of the overall fluctuation coefficient and the slope value of any muscle as a monitoring indicator; and identifying muscles whose monitoring indicators exceed a preset threshold as target muscles.

[0095] The preset index threshold is the threshold value of the monitoring index. In this embodiment of the invention, the preset index threshold can be, for example, 0.6, that is, muscles with a monitoring index greater than 0.6 are taken as target muscles.

[0096] Understandably, the steeper the slope, the worse the recovery effect. The larger the overall fluctuation coefficient, the more obvious the abnormality of the muscle is in all training sessions. Therefore, it is used as the target muscle that actually causes the recovery abnormality. Then, the mean value of the abnormality analysis index of the target muscle in the current training session is calculated, and the negative of the mean value is normalized as the degree of recovery.

[0097] In this embodiment of the invention, the degree of recovery characterizes the degree of recovery of the muscles on the abnormal side of the leg. The larger the mean value of the abnormality analysis index in the previous training, the greater the abnormality and the worse the recovery effect. Therefore, the negative value of the mean is normalized as the degree of recovery to realize the recovery analysis. Subsequently, the degree of recovery can be input into the pre-trained artificial intelligence network model, and the recovery plan or recovery measure suggestions can be output according to the target muscle and the degree of recovery.

[0098] This invention combines electrical signals from the normal and abnormal legs during training. It performs delay analysis based on the activation time of the electrical signals to determine a time deviation coefficient, and determines an intensity deviation coefficient based on the difference in activation intensity. Based on these two coefficients, the initial abnormality level of the abnormal leg is determined. Compared to related technologies that directly use interrogative recovery assessments and various discrete examination indicators, this invention combines continuous time-series analysis of data changes during exercise to construct an initial abnormality level that more closely approximates the true recovery state. Then, it analyzes the compensatory behavior of other muscles to determine the degree of abnormal compensation disorder. Combining the initial abnormality level and the degree of abnormal compensation disorder, it determines the abnormality analysis index for each muscle. This accurately captures the abnormal nature of postoperative muscle coordination, analyzes the compensatory effects of multiple muscle linkage combinations, and thus corrects the initial abnormality level, obtaining more accurate and realistic abnormality analysis indicators. Finally, it combines the changes in abnormality analysis indicators under different training sessions for rehabilitation-assisted monitoring. In summary, this invention can monitor changes in muscle electrical signals in real time and analyze them through compensatory characteristics, thereby promptly identifying subtle changes, recovery trends, and potential rehabilitation problems during the rehabilitation process and obtaining more accurate monitoring results.

[0099] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0100] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A postoperative rehabilitation monitoring system for orthopedic surgery based on artificial intelligence, characterized in that, The system includes: The acquisition module is used to acquire the electrical signals of the nerve drive of the leg muscle groups in the normal and abnormal legs under different training sessions, and to determine the activation time and activation intensity of the leg muscles at each position on each side from the electrical signals. The performance analysis module is used to perform delay analysis based on the activation time of the electrical signals of the muscles on both sides of the legs to determine the time deviation coefficient, and to determine the intensity deviation coefficient based on the difference in activation intensity of the electrical signals of the muscles on both sides of the legs. Combining the time deviation coefficient and the intensity deviation coefficient, the initial abnormal performance degree of the abnormal side leg is determined. The method for determining the initial abnormal performance degree includes calculating the mean of the time deviation coefficient and the intensity deviation coefficient as the initial abnormal performance degree of the abnormal side leg. The compensation analysis module is used to identify the adjacent muscles of each muscle at the activation time based on the activation time of different muscles on the abnormal side of the leg. It then combines the abnormal performance of the analyzed muscles to perform compensation analysis and determine the degree of abnormal compensation disorder. The method for determining the degree of abnormal compensation disorder includes calculating the mean of the abnormal performance of all analyzed muscles, normalizing the negative of the mean, and using it as the degree of abnormal compensation disorder. Finally, it combines the initial abnormal performance and the degree of abnormal compensation disorder to determine the abnormal analysis index for each muscle. The monitoring module is used to determine the functional performance change curve based on the temporal distribution of abnormal analysis indicators of each muscle in different training sessions; and to perform artificial intelligence-assisted rehabilitation monitoring based on the functional change curves of all muscles.

2. The artificial intelligence-based postoperative rehabilitation monitoring system for orthopedic surgery as described in claim 1, characterized in that, The method of determining the time deviation coefficient by performing delay analysis based on the activation time of the electrical signals of the muscles on both legs includes: The time interval between the activation time of the electrical signal of each muscle and the activation time of the first muscle is determined as the delay time of the corresponding muscle. The absolute value of the difference in the time delay of the muscles at the same location between the abnormal and normal legs is taken as the time delay degree of the corresponding muscles. The mean time delay of the muscles at each position on the abnormal side of the leg was calculated and normalized as the time deviation coefficient.

3. The artificial intelligence-based postoperative rehabilitation monitoring system for orthopedic surgery as described in claim 1, characterized in that, The determination of the intensity deviation coefficient based on the difference in activation intensity of electrical signals in the muscles of both legs includes: The absolute value of the difference in activation intensity of muscle electrical signals at the same location is used as the intensity difference of the corresponding muscle. The mean value of the muscle strength difference at each position on the abnormal side of the leg was calculated and normalized as the strength deviation coefficient.

4. The artificial intelligence-based postoperative rehabilitation monitoring system for orthopedic surgery as described in claim 1, characterized in that, The step of determining the adjacent muscles of each muscle at the activation time based on the activation time of different muscles in the abnormal leg includes: The activation sequence is obtained by sorting all muscles in the abnormal leg according to the order of activation. On the activation sequence, identify the two other muscles that are closest to any muscle sequence as the muscles to be analyzed.

5. The artificial intelligence-based postoperative rehabilitation monitoring system for orthopedic surgery as described in claim 1, characterized in that, The abnormality analysis indicators for each muscle are determined by combining the initial abnormality level and the abnormal compensation disorder level, including: The abnormal compensation disorder is linearly mapped to the value range of [0.5, 1.5] to obtain the compensation weight; The product of the initial abnormality level and the compensation weight is calculated as the anomaly analysis index.

6. The artificial intelligence-based postoperative rehabilitation monitoring system for orthopedic surgery as described in claim 1, characterized in that, The determination of functional performance change curves based on the temporal distribution of abnormal analysis indicators for each muscle across different training sessions includes: The abnormality analysis sequence is obtained by sorting the abnormality analysis indicators of each muscle in all training sessions. A two-dimensional rectangular coordinate system is constructed with the number of occurrences as the x-axis and the anomaly analysis index as the y-axis to determine the coordinates of each element in the anomaly analysis sequence in the two-dimensional rectangular coordinate system. Curve fitting was performed on all coordinate points using the least squares method to obtain the functional performance change curve.

7. The artificial intelligence-based postoperative rehabilitation monitoring system for orthopedic surgery as described in claim 1, characterized in that, The AI-assisted rehabilitation monitoring based on the functional change curves of all muscles includes: Based on the fluctuations in the functional change curves of all muscles, target muscles for key abnormality monitoring were selected. Calculate the mean of the abnormality analysis index for all target muscles in the current training session, and normalize the negative of the mean as the degree of recovery.

8. The artificial intelligence-based postoperative rehabilitation monitoring system for orthopedic surgery as described in claim 7, characterized in that, The process involves selecting target muscles for focused anomaly monitoring based on the fluctuations in the functional change curves of all muscles, including: The functional change curves of all muscles are fitted with straight lines to obtain fitted straight lines, and the slope of the fitted straight lines is determined. The average value of the ordinate of all coordinate points in the functional change curve is calculated to obtain the overall fluctuation coefficient. The product of the overall fluctuation coefficient and the slope of any muscle is normalized and used as a monitoring indicator. Muscles whose monitored indicators are greater than a preset threshold are designated as target muscles.

Images