Deep learning-based popliteal muscle strength assessment device and method

Abstract

The invention discloses a popliteal musculus strength assessment device and method based on deep learning, and relates to the technical field of sports biomechanics, the popliteal musculus strength assessment device comprises a three-dimensional force measuring table, a jacking protection device, an upper force measuring plate and a lower force measuring plate are arranged on the three-dimensional force measuring table, and the three-dimensional force measuring table further comprises an adjustable stretching device and a height adjusting locking device; the adjustable stretching device achieves distance adaptation, and the height adjusting locking device completes ankle stable fixing and height adjusting. When the device is used, equipment parameters are firstly adjusted to adapt to the body type of a subject, then force value data below knees and at ankles are synchronously collected through the double force measuring plates, safety is guaranteed in combination with protection monitoring, finally, the data are processed through the deep learning terminal, and an evaluation report containing core indexes and targeted suggestions is output. The device is high in adaptability, convenient to adjust, accurate in data, safe and practical, and suitable for popliteal muscle strength assessment of multiple people.

Description

Technical Field

[0001] This invention relates to the field of sports biomechanics, specifically to a device and method for assessing hamstring strength based on deep learning. Background Technology

[0002] Hamstring strength assessment is a core component of sports training optimization, rehabilitation progress monitoring, and lower limb injury risk prevention. The scientific validity of the assessment results directly impacts training program development and rehabilitation outcome assessment. Currently, hamstring strength assessment devices on the market are mainly divided into two categories: professional-grade high-precision equipment and simple assessment tools. Professional equipment achieves accurate assessment through multi-dimensional mechanical signal acquisition, while simple tools meet basic assessment needs with low cost and ease of operation. Both types of devices support hamstring strength testing in different scenarios.

[0003] However, existing assessment devices generally suffer from insufficient adaptability, making it difficult to match the physiological differences among different populations. Whether professional equipment or simple tools, their core force-bearing structures and the positions and spacing of fixed components are mostly fixed designs, unable to be flexibly adjusted according to individual differences such as height, leg length, and shoulder width of the subjects. For subjects who are tall or short, the fixed device structure can easily lead to non-standard force application postures, affecting not only the accuracy of data collection but also potentially causing muscle discomfort due to poor body-device compatibility. Furthermore, existing devices lack targeted adjustment mechanisms for the limb proportion differences of different age groups (such as youth and middle-aged people), further limiting their application in a wider range of populations.

[0004] Furthermore, existing devices lack sufficient ease of adjustment and flexibility, making it difficult to quickly respond to the needs of different subjects. While some devices have basic adjustment functions, the adjustment process is cumbersome, requiring the use of specialized tools to disassemble or reassemble components, which is difficult and time-consuming, reducing assessment efficiency. Other devices have limited adjustment ranges, allowing only minor positional adjustments, failing to fundamentally address the compatibility issues for subjects of different body types. These deficiencies in the adjustment mechanism result in poor adaptability of the devices in group assessment scenarios, failing to guarantee data consistency and reliability, and failing to meet the needs of ordinary users for convenient and personalized assessments, thus hindering the popularization and promotion of hamstring strength assessment technology.

[0005] Therefore, a deep learning-based hamstring strength assessment device and method are proposed to solve the above problems. Summary of the Invention

[0006] In view of this, the technical problem to be solved by the present invention is to propose a hamstring strength assessment device and method based on deep learning, which solves the problems of poor adaptability and cumbersome adjustment in the prior art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a hamstring strength assessment device and method based on deep learning, comprising a three-dimensional force measuring platform, a lifting protection device installed on one side of the three-dimensional force measuring platform, upper force measuring plates symmetrically installed on the upper surface of the three-dimensional force measuring platform, and lower force measuring plates symmetrically installed on the lower side of the three-dimensional force measuring platform; the hamstring strength assessment device based on deep learning further includes an adjustable stretching device and a height adjustment locking device.

[0008] The adjustable tensioning device is installed in the three-dimensional force measuring platform, and the adjustable tensioning device is used for force transmission control of the lower force measuring plate; The height adjustment and locking device is installed above the three-dimensional force measuring platform, and is used for convenient installation and adjustment by the operator.

[0009] Preferably, the adjustable tensioning device includes an n-shaped support rod, the bottom of which is fixedly installed on the upper surface of the lower force measuring plate. A rocker plate is movably installed on the outer surface of the middle part of the n-shaped support rod. A bearing is installed in the middle of the rocker plate. A limit rod is slidably installed on the inner surface of the bearing. The two ends of the limit rod are fixedly installed on the three-dimensional force measuring platform. Threaded sleeves are symmetrically slidably installed on the outer surface of the limit rod. An adjusting cylinder is threadedly connected to the outer surface of the threaded sleeve. The adjusting cylinder is rotatably installed in the middle of the limit rod.

[0010] Preferably, the height adjustment locking device includes a first collar, which is fixedly installed on the rocker end away from the n-shaped support rod. A second collar is engaged in the middle of the first collar. A screw is threadedly connected to the middle of the upper end of the second collar. A retaining ring is fixedly installed on the end of the screw away from the second collar. A slip ring is slidably installed in the retaining ring. A handle is fixedly installed on the outer surface of the slip ring. Arc springs are symmetrically arranged in the retaining ring. One end of the arc spring is fixedly installed on the slip ring, and the other end of the arc spring is fixedly installed in the retaining ring.

[0011] A method for using a deep learning-based hamstring strength assessment device includes the following steps: S1, Equipment parameter adjustment: By adapting and adjusting the device spacing, height, fixed parts and protective distance, a safe and comfortable testing posture is constructed for the subject, ensuring the accuracy of subsequent data collection; S2, Data Acquisition Process: Standardize the test subject's posture and force exertion movements, synchronously collect target force value data through dual force plates, and combine protective monitoring to ensure test safety, so as to achieve accurate capture and transmission of effective data; S3, Data Processing and Output: The collected data is processed and analyzed using a deep learning terminal to generate an assessment report containing core indicators and targeted suggestions, thus completing the hamstring strength assessment.

[0012] Preferably, S1 specifically includes: S1.1, Spacing Adjustment: Based on the subject's shoulder width and leg length, rotate the adjusting cylinder to drive the threaded sleeve to slide on the limiting rod, and drive the rocker to move horizontally through the bearing, adjusting the spacing between the two rockers so that when the subject stands on the three-dimensional force measuring platform, the corresponding positions of the two legs and the rocker are adapted, and the two legs are naturally separated in a comfortable posture; the core purpose is to adapt to the subject's body shape and ensure that the standing posture meets the testing standards.

[0013] S1.2, Height Adjustment: Have the subject stand at the designated position on the three-dimensional force measuring platform, and turn the screw to adjust the height of the retaining ring so that the retaining ring is level with the subject's ankle, ensuring that there is no pulling or loosening when fixing the ankle later; the core purpose is to accurately match the ankle position, lay the foundation for stable fixation, and avoid the fixation process affecting the test experience and data.

[0014] S1.3 Ankle fixation: Pull the handle to make the sliding ring slide inside the retaining ring. Place the subject's ankle between the retaining ring and the sliding ring. Release the handle. Under the elastic action of the arc spring, the sliding ring will fit tightly against the outside of the ankle, completing the stable fixation of the ankle. The fixation force should be such that it does not affect blood circulation and there is no risk of dislodgement. The core purpose is to achieve safe and stable fixation of the ankle, which can prevent displacement during the test and ensure the comfort of the subject.

[0015] S1.4, Protective Distance Setting: Adjust the initial height and horizontal distance of the lifting protective device according to the subject's height and standing posture to maintain a safe distance of 10-15cm between the protective mechanism and the subject's body, while ensuring that the sensor monitors the subject's posture changes in real time; the core purpose is to build a safety protection barrier, take into account posture monitoring, and safeguard the testing process.

[0016] Preferably, S2 specifically includes: S2.1 Initial State Acquisition: Have the subject stand with both feet on the corresponding upper force plate, body naturally upright, arms hanging naturally or holding the auxiliary handrails on both sides of the device, and maintain body stability for 30 seconds. Initial state data is collected through the upper force plate, lower force plate and sensors. The core purpose is to obtain baseline data of the subject's resting state to provide a reference for subsequent force data comparison.

[0017] S2.2, Force exertion action data collection: Issue a force exertion command and instruct the subject to perform hamstring muscle force exertion training according to the standard movements. During the force exertion, the body should remain still, and the duration of the force exertion should be controlled within 5-8 seconds. After completing one force exertion, relax for 30 seconds. The core purpose is to guide the subject to perform force exertion in a standardized manner and ensure that the collected data can truly reflect the functional status of the hamstring muscles.

[0018] S2.3, Repeated Data Acquisition and Data Transmission: Repeat the force exertion action in step 2.2 3-5 times. During each action, the upper force plate continuously collects the time series force value data Fknee(t) below the knee, and the lower force plate continuously collects the time series data Fankle(t) of the ground reaction force corresponding to the pull force at the ankle. All data are transmitted to the deep learning data processing terminal in real time. The core purpose is to improve data reliability through repeated acquisition and to synchronously complete the transmission of target data to the processing terminal.

[0019] S2.4 Safety Protection Monitoring: During the data collection process, the status of the lifting protection device is observed in real time. If the subject experiences tilting or imbalance, the protection device is immediately activated to help the subject maintain stability, and data collection is paused. Data collection is restarted after the subject regains stability. The core purpose is to avoid test safety risks in real time and ensure that the effectiveness of data collection is not affected under abnormal conditions.

[0020] Preferably, S3 specifically includes: S3.1, Data Preprocessing: The deep learning data processing terminal receives the collected Fknee(t) and Fankle(t) time series data, performs outlier removal and data standardization preprocessing on the data; the core purpose is to purify the data, eliminate interference factors, and ensure that the data input to the model meets the operation standards.

[0021] S3.2, Model Calculation and Analysis: The preprocessed data is input into the loaded deep learning model. The model extracts features and performs nonlinear mapping calculations to output the peak value of the subject's hamstring eccentric force, the average eccentric force, and the knee joint torque change curve. The core objective is to extract the core indicators of hamstring strength from the data through deep learning model calculations and complete the quantitative analysis.

[0022] S3.3, Assessment Report Generation: The terminal automatically generates an assessment report, which includes the subject's basic information, data collection parameters, core indicators of hamstring strength, comparative analysis with reference values ​​of healthy individuals, and targeted suggestions. The core purpose is to transform the analysis results into an intuitive and practical assessment report, providing clear conclusions and optimization directions for the subject's hamstring strength status.

[0023] Compared with the prior art, the present invention provides a hamstring strength assessment device and method based on deep learning, which has the following beneficial effects: (1) It has stronger adaptability and covers the physiological characteristics of diverse populations; Existing assessment devices often feature fixed core force-bearing structures and fixing components, making it difficult to adapt to individual differences in height, leg length, and shoulder width, easily leading to non-standard force application postures. This solution, through the coordinated design of an adjustable tension device and a height adjustment and locking device, allows for flexible adjustment according to the subject's body shape: rotating the adjustment cylinder adjusts the distance between the two rocker arms to accommodate different shoulder widths and leg lengths; turning the screw precisely adjusts the height of the retaining ring to match the ankle position, while the arc-spring-driven slip ring adapts to ankle thickness for stable fixation. This allows for a comfortable and suitable testing posture for individuals ranging from youth to middle age, and from tall to short, significantly expanding the device's applicability.

[0024] (2) The adjustment operation is convenient and efficient, reducing the threshold for use; Existing devices require disassembly and reassembly with specialized tools for adjustment, which is cumbersome and time-consuming; others have limited adjustment range, failing to fundamentally solve compatibility issues. This solution simplifies the adjustment logic, eliminating the need for specialized tools for all key adjustments: spacing adjustment is achieved simply by rotating the adjustment cylinder to move the rocker horizontally; height adjustment is accomplished by turning the screw; and ankle fixation is achieved quickly by pulling the handle. The entire adjustment process is intuitive and easy to understand, allowing for rapid completion by a single person. This improves efficiency in group assessment scenarios and reduces the learning curve for ordinary users, facilitating technology adoption.

[0025] (3) Data collection is accurate and reliable, ensuring the scientific nature of the assessment; Existing devices often suffer from poor adaptability, leading to abnormal force application postures and affecting data accuracy. Furthermore, most devices only collect data from a single dimension. This solution simultaneously collects force data (Fknee(t) below the knee) and ground reaction force data (Fankle(t) at the ankle) using dual force plates. Combined with standardized force application guidance and a 3-5 repetition acquisition mechanism, this ensures the data accurately reflects the functional state of the hamstrings. Simultaneously, the initial state data acquisition provides a precise reference for force application data. In the preprocessing stage, outliers are removed and data is standardized. Then, a deep learning model is used for feature extraction and quantitative analysis, making the evaluation results more scientific and reliable.

[0026] (4) Comprehensive safety protection, taking into account both testing experience and risk control; Existing devices generally lack specific safety features, and subjects may face the risk of falling due to force imbalance or loose fixation. This solution incorporates a lifting safety device that uses sensors to monitor the subject's posture in real time, maintaining a safe distance of 10-15cm from the body. In the event of tilting, imbalance, or other abnormalities, an auxiliary stabilization mechanism can be activated immediately. Simultaneously, the height adjustment locking device uses an arc-shaped spring to elastically secure the ankle, avoiding excessive tightness that could impair blood circulation and preventing dislodgement during testing. This minimizes discomfort to the subject from the fixation components while ensuring testing safety, thus improving the testing experience.

[0027] (5) Intelligent analysis and practical output enhance the evaluation and guidance value; Existing devices mostly output basic force data, lacking in-depth analysis and targeted suggestions, making it difficult for users to quickly obtain core information. This solution integrates deep learning technology, inputting collected time-series data into a trained model, which can automatically output core indicators such as peak hamstring eccentric force, average eccentric force, and knee joint torque variation curves. The final evaluation report not only includes data comparison and analysis but also provides targeted suggestions such as adjusting training intensity and strengthening weak areas tailored to the individual, transforming data into directly applicable guidance plans and significantly enhancing the practical value of the evaluation. Attached Figure Description

[0028] Figure 1 This is a three-dimensional structural diagram of a hamstring strength assessment device based on deep learning according to the present invention. Figure 2 This is a three-dimensional structural auxiliary diagram of a hamstring strength assessment device based on deep learning according to the present invention. Figure 3 For the present invention Figure 3 Enlarged view of point A in the middle; Figure 4 This is a schematic diagram of the structural connection relationship of the adjustable tensioning device of the present invention; Figure 5 For the present invention Figure 5 Enlarged view of section B in the middle.

[0029] In the picture: 1. Three-dimensional force measuring platform; 11. Lifting and protective device; 12. Upper force measuring plate; 13. Lower force measuring plate; 2. Adjustable tensioning device; 21. N-shaped support rod; 22. Rocker; 23. Limiting rod; 24. Bearing; 25. Threaded sleeve; 26. Adjusting cylinder; 3. Height adjustment locking device; 31. First collar; 32. Second collar; 33. Screw; 34. Snap ring; 35. Slip ring; 36. Curved spring; 37. Pull handle. Detailed Implementation

[0030] The technical solutions of the embodiments 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, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0032] Example 1, please refer to Figures 1 to 5 As shown: To address the problems mentioned in the technical solutions, this application provides a deep learning-based hamstring strength assessment device, specifically comprising: A hamstring strength assessment device based on deep learning includes a three-dimensional force measuring platform 1, a lifting protection device 11 installed on one side of the three-dimensional force measuring platform 1, an upper force measuring plate 12 symmetrically installed on the upper surface of the three-dimensional force measuring platform 1, and a lower force measuring plate 13 symmetrically installed on the lower side of the three-dimensional force measuring platform 1. The hamstring strength assessment device based on deep learning also includes an adjustable stretching device 2 and a height adjustment locking device 3. An adjustable tensioning device 2 is installed in the three-dimensional force measuring platform 1. The adjustable tensioning device 2 is used for force transmission control of the lower force measuring plate 13. The height adjustment and locking device 3 is installed above the three-dimensional force measuring table 1. The height adjustment and locking device 3 is for the convenient installation by the operator. Specifically, the bottom of the n-shaped support rod 21 is fixedly installed on the upper surface of the lower force measuring plate 13. A rocker plate 22 is movably installed on the outer surface of the middle part of the n-shaped support rod 21. A bearing 24 is installed in the middle of the rocker plate 22. A limit rod 23 is slidably installed on the inner surface of the bearing 24. The two ends of the limit rod 23 are fixedly installed on the three-dimensional force measuring table 1. A threaded sleeve 25 is symmetrically slidably installed on the outer surface of the limit rod 23. An adjusting cylinder 26 is threadedly connected to the outer surface of the threaded sleeve 25. The adjusting cylinder 26 is rotatably installed in the middle of the limit rod 23. Among them, the upper force plate 12 consists of two force plates installed on the upper surface of the three-dimensional force platform 1, which are used to measure and collect the time series below the knee as Fknee(t), and the lower force plate 13 consists of two force plates below the three-dimensional force platform 1, which are used to collect the reaction force of the subject's ankle on the ground as Fankel(t). The adjustable tension device 2 is used to convert the tension at the ankle of the subject into a reaction force of the lower force plate 13 on the ground, with the assistance of the height adjustment locking device 3 and the support of the adjustable tension device 2. At the same time, this scheme can drive the threaded sleeve 25 to slide on the limiting rod 23 by manually rotating the adjusting cylinder 26. The sliding of the threaded sleeve 25 can drive the bearing 24 to slide on the limiting rod 23. The rotation of the bearing 24 and the rocker 22 can adjust the distance between the subject's legs, so that experiments can be quickly conducted on different subjects, thereby increasing the applicability of this scheme. At the same time, this scheme is simple and convenient to adjust. Simply rotating the adjusting cylinder 26 can drive the horizontal adjustment of the rocker 22, thereby reducing the difficulty of experimental operation. In addition, this scheme has a lifting protection device 11, and the sensors installed on the lifting protection device 11 can maintain a safe distance from the subject at all times to prevent the subject from falling due to unstable force, thereby increasing the safety of the subject.

[0033] Specifically, the first ring 31 is fixedly installed on the rocker 22 at the end away from the n-shaped support rod 21. The second ring 32 is snapped into the middle of the first ring 31. The upper middle of the second ring 32 is threaded with a screw 33. A retaining ring 34 is fixedly installed at the end of the screw 33 away from the second ring 32. A slip ring 35 is slidably installed in the retaining ring 34. A handle 37 is fixedly installed on the outer surface of the slip ring 35. Arc springs 36 are symmetrically arranged in the retaining ring 34. One end of the arc spring 36 is fixedly installed on the slip ring 35, and the other end of the arc spring 36 is fixedly installed in the retaining ring 34. This solution allows the retaining ring 34 to adapt to different installation heights by setting the thread adjustment of the second ring 32 and the screw 33. At the same time, by the operator gently pulling the handle 37, the sliding ring 35 can slide in the retaining ring 34 to wrap the outside of the subject's ankle, thereby preventing the subject from falling and getting injured due to unstable ankle installation.

[0034] Example 2 differs from Example 1 in that it provides a deep learning-based hamstring strength assessment device, the specific operation of which is as follows: (a) Equipment inspection and debugging; 1. Core component inspection: Confirm that the upper force plate 12 and lower force plate 13 of the three-dimensional force measuring platform 1 are undamaged and free of stains, and that the data acquisition module responds normally after power-on and can output force value data in real time; check the sensor sensitivity of the lifting protection device 11 to ensure that it can accurately detect distance and that the protection mechanism can be activated flexibly.

[0035] 2. Adjustable tensioning device debugging: Rotate the adjusting cylinder 26 and observe whether the threaded sleeve 25 slides smoothly on the limiting rod 23, whether the rocker plate 22 rotates flexibly through the bearing 24, and whether the fixed connection between the n-shaped support rod 21 and the lower force measuring plate 13 is loose.

[0036] 3. Height adjustment locking device debugging: Tighten the screw 33 to verify whether the thread fit between the second ring 32 and the screw 33 is reliable. The arc spring 36 in the retaining ring 34 has normal elasticity. When the handle 37 is pulled, the slip ring 35 slides freely and can quickly reset after being released.

[0037] 4. Data transmission debugging: Connect the three-dimensional force measuring platform to the deep learning data processing terminal to ensure that the force value data Fknee(t) below the knee collected by the upper force measuring plate 12 and the ground reaction force data Fankle(t) at the ankle collected by the lower force measuring plate 13 can be transmitted to the terminal in real time without delay or loss.

[0038] (ii) Subject preparation; 1. Collect basic information about the subjects, including age, gender, height, weight, exercise habits, whether they have a long-term exercise history, hamstring training history, and no history of lower limb injury.

[0039] 2. Have the subjects change into tight-fitting sportswear and non-slip sports shoes, and remove jewelry and heavy clothing from the ankles and calves to avoid interfering with equipment fixation and data collection.

[0040] 3. Explain the assessment process and proper procedures to the subjects, such as body posture, force application, and safety precautions during the test. Inform the subjects that they can signal to stop immediately if they experience any discomfort.

[0041] (III) Preprocessing of deep learning models; 1. Load the pre-trained deep learning model for hamstring strength assessment. The model is trained on a large amount of sample data. The input is Fknee(t) and Fankle(t) time series data, and the output is hamstring eccentric force and knee joint torque.

[0042] 2. Configure the model input data format to ensure that the time series data collected by the force measuring table can directly match the model input requirements without additional format conversion.

[0043] The specific implementation steps are as follows: (a) Equipment parameter adjustment 1. Spacing adjustment: Based on the subject's shoulder width and leg length, rotate the adjusting cylinder 26 to drive the threaded sleeve 25 to slide on the limiting rod 23, and drive the rocker 22 to move horizontally through the bearing 24, adjusting the spacing between the two rocker 22 so that when the subject stands on the three-dimensional force measuring platform 1, the position of the two legs is adapted to the corresponding position of the rocker 22, and the two legs are naturally separated in a comfortable posture.

[0044] 2. Height adjustment: Have the subject stand at the designated position on the three-dimensional force measuring platform 1, and turn the screw 33 to adjust the height of the retaining ring 34 so that the retaining ring 34 is level with the subject's ankle, ensuring that there is no pulling or loosening when fixing the ankle later.

[0045] 3. Ankle fixation: Pull the handle 37 to make the sliding ring 35 slide inside the retaining ring 34. Place the subject's ankle between the retaining ring 34 and the sliding ring 35. Release the handle 37. Under the elastic action of the arc spring 36, the sliding ring 35 will fit tightly against the outside of the ankle, completing the stable fixation of the ankle. Ensure that the fixation force is moderate, does not affect blood circulation, and has no risk of falling off.

[0046] 4. Protective distance setting: Adjust the initial height and horizontal distance of the lifting protective device 11 according to the subject's height and standing posture to maintain a safe distance of 10-15cm between the protective mechanism and the subject's body, while ensuring that the sensor monitors the subject's posture changes in real time.

[0047] (ii) Data acquisition process; 1. Have the subject stand with both feet on the corresponding upper force plate 12, body naturally upright, hands hanging down naturally or holding the auxiliary handrails on both sides of the device if available, and keep the body stable for 30 seconds as the initial state data collection.

[0048] 2. Issue a force exertion command and have the subject perform hamstring muscle exertion training according to the standard movements, such as seated leg curls, standing eccentric hamstring contractions, etc. Keep the body still during the exertion, control the duration of the exertion to 5-8 seconds, and relax for 30 seconds after completing one exertion.

[0049] 3. Repeat the force application action 3-5 times. During each action, the upper force plate 12 continuously collects the time series force value data Fknee(t) below the knee, and the lower force plate 13 continuously collects the time series data Fankle(t) of the ground reaction force corresponding to the pull force at the ankle. All data are transmitted to the deep learning data processing terminal in real time.

[0050] 4. During the data collection process, observe the status of the lifting protection device 11 in real time. If the subject experiences tilting or imbalance, the protection device will be activated immediately to help the subject maintain stability. At the same time, data collection will be paused and resumed after the subject has regained stability.

[0051] (III) Data Processing and Result Output 1. The deep learning data processing terminal receives the collected Fknee(t) and Fankle(t) time series data and performs preprocessing on the data (including outlier removal and data standardization).

[0052] 2. Input the preprocessed data into the loaded deep learning model. The model outputs the peak value of the hamstring eccentric force, the average eccentric force, and the knee joint torque change curve of the subject through feature extraction, nonlinear mapping and other operations.

[0053] 3. The terminal automatically generates an assessment report, which includes the subject's basic information, data collection parameters, core indicators of hamstring strength, comparative analysis with reference values ​​of healthy individuals, and targeted suggestions (such as adjustments to training intensity and directions for strengthening weak areas).

[0054] This program includes a detailed statistical table of participant data collection.

[0055] 1. Complete the assessment data collection for the two groups of subjects according to the above parameter settings, and record the Fknee(t) and Fangle(t) data of each force exertion of each group of subjects, as well as the hamstring eccentric force and knee joint torque index output by the model.

[0056] 2. The average peak eccentric hamstring force of young men and middle-aged men was approximately 280-320N and 180-220N for young women; while that of middle-aged men was approximately 220-260N and that of middle-aged women was approximately 140-180N, which is consistent with the muscle strength distribution pattern of different age groups.

[0057] 3. Comparative analysis of the stability of indicators for repeated actions by the same subject showed that the coefficient of variation was ≤5%, indicating that the data acquisition accuracy of the equipment was reliable and the prediction results of the deep learning model were consistent.

[0058] 4. For middle-aged participants with low exercise frequency, the assessment report recommends increasing the frequency of hamstring resistance training, with the initial training intensity controlled at 40-50% of their maximum strength, and gradually increasing it.

[0059] Please refer to the above work process. Figures 1 to 5 .

[0060] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0061] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A deep learning-based hamstring strength assessment device, characterized in that, The device includes a three-dimensional force measuring platform (1), a lifting protection device (11) installed on one side of the three-dimensional force measuring platform (1), an upper force measuring plate (12) symmetrically installed on the upper surface of the three-dimensional force measuring platform (1), and a lower force measuring plate (13) symmetrically installed on the lower side of the three-dimensional force measuring platform (1). The hamstring strength assessment device based on deep learning also includes an adjustable stretching device (2) and a height adjustment and locking device (3). The adjustable tensioning device (2) is installed in the three-dimensional force measuring platform (1), and the adjustable tensioning device (2) is used for the force transmission control of the lower force measuring plate (13); The height adjustment and locking device (3) is set above the three-dimensional force measuring table (1), and the height adjustment and locking device (3) is used for convenient installation and adjustment by the operator.

2. The hamstring strength assessment device based on deep learning according to claim 1, characterized in that: The adjustable tensioning device (2) includes an n-shaped support rod (21), the bottom of which is fixedly installed on the upper surface of the lower force measuring plate (13). A rocker plate (22) is movably installed on the outer surface of the middle part of the n-shaped support rod (21), and a bearing (24) is installed in the middle part of the rocker plate (22).

3. The hamstring strength assessment device based on deep learning according to claim 2, characterized in that: The bearing (24) has a limit rod (23) slidably mounted on its inner surface. The two ends of the limit rod (23) are fixedly mounted on the three-dimensional force measuring table (1). The outer surface of the limit rod (23) is symmetrically slidably mounted with a threaded sleeve (25). The outer surface of the threaded sleeve (25) is threadedly connected with an adjusting cylinder (26). The adjusting cylinder (26) is rotatably mounted in the middle of the limit rod (23).

4. The hamstring strength assessment device based on deep learning according to claim 3, characterized in that: The height adjustment locking device (3) includes a first collar (31), which is fixedly installed on the rocker (22) at the end away from the n-shaped support rod (21), and a second collar (32) is engaged in the middle of the first collar (31).

5. The hamstring strength assessment device based on deep learning according to claim 4, characterized in that: The upper middle part of the second ring (32) is threaded with a screw (33), and a retaining ring (34) is fixedly installed at the end of the screw (33) away from the second ring (32). A slip ring (35) is slidably installed in the retaining ring (34), and a handle (37) is fixedly installed on the outer surface of the slip ring (35).

6. The hamstring strength assessment device based on deep learning according to claim 5, characterized in that: A curved spring (36) is symmetrically arranged in the retaining ring (34). One end of the curved spring (36) is fixedly installed on the slip ring (35), and the other end of the curved spring (36) is fixedly installed in the retaining ring (34).

7. A method of using a deep learning-based hamstring strength assessment device, adapted to the deep learning-based hamstring strength assessment device according to any one of claims 1-6, characterized in that, Includes the following steps: S1, Equipment parameter adjustment: By adapting and adjusting the device spacing, height, fixed parts and protective distance, a safe and comfortable testing posture is constructed for the subject, ensuring the accuracy of subsequent data collection; S2, Data Acquisition Process: Standardize the test subject's posture and force exertion movements, synchronously collect target force value data through dual force plates, and combine protective monitoring to ensure test safety, so as to achieve accurate capture and transmission of effective data; S3, Data Processing and Output: The collected data is processed and analyzed using a deep learning terminal to generate an assessment report containing core indicators and targeted suggestions, thus completing the hamstring strength assessment.

8. The method of using the deep learning-based hamstring strength assessment device according to claim 7, characterized in that, S1 specifically includes: S1.1, Spacing adjustment: Based on the subject's shoulder width and leg length, rotate the adjusting cylinder (26) to drive the threaded sleeve (25) to slide on the limiting rod (23), and drive the rocker (22) to move horizontally through the bearing (24) to adjust the spacing between the two rocker (22) so that when the subject stands on the three-dimensional force measuring platform (1), the corresponding position of the two legs is matched with the rocker (22), and the two legs are naturally separated in a comfortable posture; the core purpose is to match the subject's body shape and ensure that the standing posture meets the test standards; S1.2, Height Adjustment: Have the subject stand at the designated position on the three-dimensional force measuring platform (1), turn the screw (33) to adjust the height of the retaining ring (34) so ​​that the retaining ring (34) is level with the subject's ankle position, ensuring that there is no pulling or loosening when fixing the ankle later; the core purpose is to accurately match the ankle position, lay the foundation for stable fixation, and avoid the fixation process affecting the test experience and data. S1.3, Ankle fixation: Pull the handle (37) to make the sliding ring (35) slide inside the retaining ring (34), place the subject's ankle between the retaining ring (34) and the sliding ring (35), release the handle (37), and under the elastic action of the arc spring (36), the sliding ring (35) will be tightly attached to the outside of the ankle, thus completing the stable fixation of the ankle. The fixation force should be such that it does not affect blood circulation and there is no risk of falling off. The core purpose is to achieve safe and stable fixation of the ankle, which can prevent displacement during the test and ensure the comfort of the subject. S1.4, Protective distance setting: Adjust the initial height and horizontal distance of the lifting protective device (11) according to the height and standing posture of the subject, so that the protective mechanism and the subject's body maintain a safe distance of 10-15cm, while ensuring that the sensor monitors the subject's posture changes in real time; the core purpose is to build a safety protection barrier, take into account posture monitoring, and protect the test process.

9. The method of using the deep learning-based hamstring strength assessment device according to claim 7, characterized in that, S2 specifically includes: S2.1, Initial state acquisition: Have the subject stand with both feet on the corresponding upper force plate (12), body naturally straight, hands hanging down or holding the auxiliary handrails on both sides of the device, and keep the body stable for 30 seconds. Initial state data are collected through the upper force plate (12), lower force plate (13) and sensors. The core purpose is to obtain the subject's resting state baseline data to provide a reference for subsequent force data comparison. S2.2, Force exertion action data collection: Issue a force exertion command and instruct the subject to perform hamstring muscle force exertion training according to the standard movements. During the force exertion, the body should remain still, and the duration of the force exertion should be controlled within 5-8 seconds. After completing one force exertion, relax for 30 seconds. The core purpose is to guide the subject to perform force exertion in a standardized manner and ensure that the collected data can accurately reflect the functional status of the hamstring muscles. S2.3, Repeated data acquisition and data transmission: Repeat the force exertion action in step 2.2 3-5 times. During each action, the upper force plate (12) continuously acquires the time series force value data Fknee(t) below the knee, and the lower force plate (13) continuously acquires the time series data Fankle(t) of the ground reaction force corresponding to the pull force at the ankle. All data are transmitted to the deep learning data processing terminal in real time. The core purpose is to improve the reliability of the data through repeated acquisition and to synchronously complete the transmission of the target data to the processing terminal. S2.4 Safety Protection Monitoring: During the data collection process, the status of the lifting protection device (11) is observed in real time. If the subject experiences body tilting or imbalance, the protection device is immediately activated to help the subject maintain stability. At the same time, data collection is suspended and restarted after the subject recovers stability. The core purpose is to avoid test safety risks in real time and ensure that the effectiveness of data collection is not affected under abnormal conditions.

10. The method of using the deep learning-based hamstring strength assessment device according to claim 7, characterized in that, S3 specifically includes: S3.1, Data Preprocessing: The deep learning data processing terminal receives the collected Fknee(t) and Fankle(t) time series data, performs outlier removal and data standardization preprocessing on the data; the core purpose is to clean the data, eliminate interference factors, and ensure that the data input to the model meets the operation standards. S3.2, Model Calculation and Analysis: The preprocessed data is input into the loaded deep learning model. The model extracts features and performs nonlinear mapping calculations to output the peak value, average eccentric force, and knee joint torque change curve of the subject's hamstring muscles. The core objective is to extract the core indicators of hamstring strength from the data through deep learning model calculations and complete the quantitative analysis. S3.3, Assessment Report Generation: The terminal automatically generates an assessment report, which includes the subject's basic information, data collection parameters, core indicators of hamstring strength, comparative analysis with reference values ​​of healthy individuals, and targeted suggestions. The core purpose is to transform the analysis results into an intuitive and practical assessment report, providing clear conclusions and optimization directions for the subject's hamstring strength status.

Images