Intelligent scoliosis rehabilitation training feedback system

Abstract

The invention relates to the technical field of intelligent medical rehabilitation, and discloses an intelligent scoliosis rehabilitation training feedback system which comprises a mark point space coordinate capturing module used for dynamically capturing three-dimensional coordinates of mark points on the body surface of a patient; the active marker group module comprises but is not limited to a seventh cervical vertebra marker, a thoracic vertebra convex side marker, a lumbar vertebra convex side marker, a left posterior superior spine iliac marker and a right posterior superior spine iliac marker, and the markers are marked at specific positions of the body surface of the patient. According to the method, three-dimensional coordinates of key mark points on the body surface of a patient are captured in real time through a mark point space coordinate capturing module, then a standard human body sagittal plane constructed based on the sacrum center line is calculated in real time through a plane calculation formula, and then the distance from the key mark points to the constructed median sagittal plane is measured in real time through a point-to-plane distance formula; and a real-time signal is generated.

Description

Technical Field

[0001] The present invention relates to the field of intelligent medical rehabilitation technology, and in particular to an intelligent scoliosis rehabilitation training feedback system. Background Art

[0002] Scoliosis is a common three-dimensional spinal deformity characterized by abnormal curvature of the spine in the coronal, sagittal, and transverse planes. It is particularly prevalent among adolescents. Adolescent idiopathic scoliosis (AIS) is the main type of scoliosis. While its etiology remains unclear, early intervention and rehabilitation training are crucial for improving spinal morphology and slowing disease progression.

[0003] Traditional scoliosis rehabilitation training methods primarily rely on manual correction by physical therapists and conventional equipment training, such as the Schroth Method and spinal braces. For example, Chinese patent document CN221180747U includes an angle adjustment mechanism and a front armrest. The angle adjustment mechanism adjusts the angle of the bed to provide varying degrees of traction correction for the patient. The height and position of the front armrest can be adjusted according to the user's needs, making it easier for the patient to grasp. Another example is Chinese patent document CN113052842A, which uses trained segmentation and regression networks to output more accurate predictions of scoliosis angles. While these methods can improve scoliosis to a certain extent, they still have the following problems:

[0004] Highly subjective: The training effect depends on the therapist's experience and judgment, lacks objective data support, and is difficult to quantify training progress;

[0005] Lack of real-time feedback: Patients cannot understand the deviation between their own posture and the target posture in real time, resulting in low training efficiency;

[0006] Lack of personalization: The training program is difficult to dynamically adjust according to the individual differences of patients and cannot meet the rehabilitation needs of different patients. Summary of the Invention

[0007] (1) Technical problems solved

[0008] In response to the deficiencies of the prior art, the present invention provides an intelligent scoliosis rehabilitation training feedback system, which solves the problems of the above-mentioned background technology of "strong subjectivity, lack of real-time feedback and lack of personalization".

[0009] (2) Technical solution

[0010] To achieve the above objectives, the present invention is implemented through the following technical solutions: an intelligent scoliosis rehabilitation training feedback system, comprising:

[0011] A marker point spatial coordinate capture module, which is used to dynamically capture the three-dimensional coordinates of the patient's body surface marker points;

[0012] An active marker group module, including but not limited to the seventh cervical vertebra marker, the thoracic vertebra convex side marker, the lumbar vertebra convex side marker, the left posterior superior iliac spine marker, and the right posterior superior iliac spine marker, which are marked at specific locations on the patient's body surface;

[0013] The dynamic sagittal plane construction module constructs the standard human anterior and posterior median sagittal plane based on the sacral midline, which serves as an important reference plane for rehabilitation training;

[0014] A real-time feedback module, configured to generate a real-time feedback signal according to the distance between the calculated marker point and the constructed sagittal plane;

[0015] A display module is used to display the captured marker information, dynamic sagittal plane construction results and real-time feedback signals on the screen;

[0016] Among them, the goal of the rehabilitation training is to bring the locations of specific marker points (seventh cervical vertebra marker, thoracic convex side marker, lumbar convex side marker) as close as possible to the sagittal plane. In addition, the marker point spatial coordinate capture module, active marker group module, dynamic sagittal plane construction module, real-time feedback module and display module are connected through network communication.

[0017] Preferably, the marker point spatial coordinate capture module includes multiple synchronized industrial cameras, configured with high frame rate and high resolution. The marker point spatial coordinate capture module includes a synchronization unit, a calibration unit and a multi-frame fusion unit. The synchronization unit is used to ensure the synchronization of multiple industrial cameras in time and space, and capture the three-dimensional coordinates of the marker points with high precision. The calibration unit is used to regularly perform spatial calibration on the camera to ensure capture accuracy. The multi-frame fusion unit is used to fuse the data captured by multiple cameras to improve the accuracy of the three-dimensional coordinates. The marker point spatial coordinate capture module also includes a laser radar unit. The laser radar unit and the industrial camera group constitute a multi-modal sensor array, which compensates for the blind spot of the optical camera's field of view through pulse time difference ranging.

[0018] Preferably, the active marker group module includes a high contrast unit, a wireless communication unit, an adjustable unit and a reflective material unit. The high contrast unit is mainly used to enable the marker to maintain high contrast under different lighting conditions to facilitate camera capture. The wireless communication unit is used to transmit the status information of the marker to the system in real time. The adjustable unit is used to adjust the position and size of the marker according to the patient's body shape to ensure the accuracy of the marking. The reflective material unit is used to enhance the visibility of the marker.

[0019] Preferably, the dynamic sagittal plane construction module includes a machine learning unit, an adaptive unit, and a historical data unit. The machine learning unit is used to construct and optimize the human sagittal plane in real time based on the captured marker point data. The adaptive unit is used to dynamically adjust the sagittal plane construction parameters according to the patient's body shape and posture changes. The historical data unit is used to store and analyze historical construction data and optimize the sagittal plane construction algorithm. The sagittal plane calculation method is as follows:

[0020] Assume that there are two points in space: the left posterior superior iliac spine A (x1, y1, z1) and the right posterior superior iliac spine B (x2, y2, z2). First, find the coordinates of the midpoint M of AB: According to the midpoint coordinate formula, the coordinates of M are

[0021] Then find the vector

[0022] Since the plane we are looking for passes through point M and is the normal vector (the plane is perpendicular to the line segment AB, then the direction vector of the line segment AB is the normal vector of the plane). According to the point normal equation of the plane A (xx m )+B(yy m )+C(zz m )=0,(where(x m ,y m , z m ) is a point on the plane, (A, B, C) is the normal vector of the plane);

[0023] Here: A=x2-x1, B=y2-y1, C=z2-z1;

[0024] Then the equation of the plane is

[0025] In addition, a specific marker point P(x0,y0.z 0) The calculation formula for the distance d to the required sagittal plane is:

[0026]

[0027] Preferably, the real-time feedback module includes a dynamic adjustment unit, a multimodal feedback unit and a threshold setting unit. The dynamic adjustment unit is used to dynamically adjust the intensity and frequency of the feedback signal according to the progress of training. The multimodal feedback unit is used to provide multiple feedback methods such as vision, hearing and touch to enhance the patient's perception. The threshold setting unit is used to set the trigger threshold of the feedback signal according to the specific situation of the patient.

[0028] Preferably, the display module includes a visualization unit, a multi-view display unit and a data analysis unit, and the visualization unit is used to display the distance between the marking point and the sagittal plane in a graphical manner on the screen in real time.

[0029] The multi-view display unit is used to display the positional relationship between the marker points and the sagittal plane from different viewpoints, and the data analysis unit is used to display the training data in the form of a chart to facilitate patients and therapists to understand the progress of training.

[0030] Preferably, the system further comprises an alarm module, which is used to send out an alarm signal when the distance between the marking point and the sagittal plane exceeds a preset threshold, and can also provide alarm signals of different levels according to the degree of distance excess.

[0031] Preferably, the synchronization error between the cameras is controlled within ±0.1ms, the calibration accuracy error is ≤0.5mm, the markers in the active marker group module can maintain a contrast ≥80% in the light intensity range of 50lux to 1000lux, the wireless communication delay is ≤10ms, and the marker position adjustment accuracy is ≤1mm.

[0032] Preferably, the sagittal plane construction error in the dynamic sagittal plane construction module needs to be ≤2mm, the machine learning unit training data set contains ≥1000 patient data, and the adaptive unit adjustment response time is ≤0.5s.

[0033] Preferably, the intensity range of the multimodal feedback unit is 0% to 100%, the frequency range of the dynamic adjustment unit is 0.1 Hz to 10 Hz, and the threshold setting unit sets the threshold range to 1 mm to 20 mm.

[0034] (3) Beneficial effects

[0035] The present invention provides an intelligent scoliosis rehabilitation training feedback system, which has the following beneficial effects:

[0036] (1) When the intelligent scoliosis rehabilitation training feedback system is in use, the three-dimensional coordinates of the key marker points on the patient's body surface are captured in real time through the marker point spatial coordinate capture module. Then, using the plane calculation formula, the standard human sagittal plane constructed based on the sacral midline is calculated in real time. The point-to-plane distance formula is then used to measure the distance from the key marker point to the constructed midsagittal plane in real time and generate a real-time signal. Through the three-dimensional coordinate fusion formula and the sagittal plane construction formula, the system can accurately calculate the distance between the marker point and the sagittal plane, generate objective training data, and avoid the limitations of traditional methods that rely on the therapist's subjective judgment.

[0037] (2) When in use, the intelligent scoliosis rehabilitation training feedback system generates real-time feedback signals based on the distance between the marker point and the sagittal plane through a dynamic adjustment unit, a multimodal feedback unit, and a threshold setting unit, and provides multiple feedback methods such as vision, hearing, and touch. This allows patients to understand the deviation between their own posture and the target posture in real time and adjust their training posture in a timely manner, significantly improving training efficiency.

[0038] (3) When in use, the intelligent scoliosis rehabilitation training feedback system dynamically adjusts the sagittal plane construction parameters according to the patient's body shape, posture changes and training progress through machine learning units, adaptive units and historical data units, optimizes the training plan, sets the trigger threshold of the feedback signal according to the patient's specific situation, and provides multimodal feedback methods to achieve personalized rehabilitation training. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 This is a schematic diagram of the overall system framework of the present invention;

[0040] Figure 2 Schematic diagram of the detailed framework of the marker point spatial coordinate capture module in the present invention;

[0041] Figure 3 Schematic diagram of the detailed framework of the active marker group module in the present invention;

[0042] Figure 4 Schematic diagram of the detailed framework of the dynamic sagittal plane building module in the present invention;

[0043] Figure 5 Detailed framework diagram of the real-time feedback module in the present invention;

[0044] Figure 6 Schematic diagram of the detailed framework of the display module in the present invention;

[0045] Figure 7 Schematic diagram of the seventh cervical vertebra marking, thoracic vertebra convex side marking, and lumbar vertebra convex side marking in the present invention;

[0046] Figure 8This is a schematic diagram of the standard human anterior and posterior sagittal plane constructed based on the sacral midline in the present invention.

[0047] In the figure: 1. Marker point spatial coordinate capture module; 101. Synchronization unit; 102. Calibration unit; 103. Multi-frame fusion unit; 2. Active marker group module; 201. High contrast unit; 202. Wireless communication unit; 203. Adjustable unit; 204. Reflective material unit; 3. Dynamic sagittal plane construction module; 301. Machine learning unit; 302. Adaptive unit; 303. Historical data unit; 4. Real-time feedback module; 401. Dynamic adjustment unit; 402. Multimodal feedback unit; 403. Threshold setting unit; 5. Display module; 501. Visualization unit; 502. Multi-view display unit; 503. Data analysis unit; 6. Alarm module. DETAILED DESCRIPTION

[0048] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0049] Example 1

[0050] See also Figure 1 - Figure 8 The present invention provides an intelligent scoliosis rehabilitation training feedback system, comprising: a marker point spatial coordinate capture module 1, an active marker group module 2, a dynamic sagittal plane construction module 3, a real-time feedback module 4, a display module 5, and an alarm module 6. The goal of rehabilitation training is to bring the location of specific marker points (seventh cervical vertebra marker, thoracic convex side marker, lumbar convex side marker) as close as possible to the sagittal plane. The marker point spatial coordinate capture module 1, the active marker group module 2, the dynamic sagittal plane construction module 3, the real-time feedback module 4, and the display module 5 are connected via network communication. The marker point spatial coordinate capture module 1 is used to dynamically capture the three-dimensional coordinates of the patient's body surface markers. The three-dimensional coordinate fusion formula is:

[0051]

[0052] Among them, P(x,y,z) is the fused three-dimensional coordinate, Pi(xi,yi,zi) is the coordinate captured by the i-th camera, and n is the number of cameras.

[0053] Specifically, the marker point spatial coordinate capture module 1 includes multiple synchronized industrial cameras, using a high frame rate, the frame rate needs to be ≥120fps and high resolution, the requirement is ≥4K, the marker point spatial coordinate capture module 1 includes a synchronization unit 101, a calibration unit 102 and a multi-frame fusion unit 103, the synchronization unit 101 is used to ensure the synchronization of multiple industrial cameras in time and space, and capture the three-dimensional coordinates of the marker points with high precision, the calibration unit 102 is used to regularly perform spatial calibration on the camera to ensure capture accuracy, and the multi-frame fusion unit 103 is used to fuse the data captured by multiple cameras to improve the accuracy of the three-dimensional coordinates. In addition, the marker point spatial coordinate capture module 1 also includes a laser radar unit 104, and the laser radar unit 104 and the industrial camera group constitute a multi-modal sensor array, which compensates for the blind spot of the optical camera's field of view through pulse time difference ranging.

[0054] The active marker group module 2 includes but is not limited to the seventh cervical vertebra marker, the thoracic vertebra convex side marker, the lumbar vertebra convex side marker, the left posterior superior iliac spine marker, and the right posterior superior iliac spine marker. These markers are marked at specific locations on the patient's body surface. Specifically, the active marker group module 2 includes a high-contrast unit 201, a wireless communication unit 202, an adjustable unit 203, and a reflective material unit 204. The high-contrast unit 201 is mainly used to enable the marker to maintain high contrast under different lighting conditions to facilitate camera capture. The wireless communication unit 202 is used to transmit the status information of the marker to the system in real time. The adjustable unit 203 is used to adjust the position and size of the marker according to the patient's body shape to ensure the accuracy of the marking. The reflective material unit 204 is used to enhance the visibility of the marker.

[0055] Dynamic sagittal plane construction module 3 constructs a standard human anterior and posterior median sagittal plane based on the sacral midline, measures the distance from key landmarks to the constructed median sagittal plane, and generates real-time signals. The sagittal plane calculation method is as follows:

[0056] Assume that there are two points in space: the left posterior superior iliac spine A (x1, y1, z1) and the right posterior superior iliac spine B (x2, y2, z2) (refer to Figure 8 ), first find the coordinates of the midpoint M of AB: According to the midpoint coordinate formula, the coordinates of M are

[0057] Then find the vector

[0058] Since the plane we are looking for passes through point M and is the normal vector (the plane is perpendicular to the line segment AB, then the direction vector of the line segment AB is the normal vector of the plane). According to the point normal equation of the plane A (xx m )+B(yy m )+C(zz m )=0,(where(xm ,y m , z m ) is a point M on the plane, (A, B, C) is the normal vector of the plane);

[0059] Here:

[0060] A=x2-x1, B=y2-y1, C=z2-z1;

[0061] Then the equation of the plane is

[0062] For example, if A(1,0,0), B(3,0,0), then the coordinates of the midpoint M are

[0063] According to the point formula equation, the plane equation is: 2(x-2)+0(y-0)+0(z-0)=0, that is, x=2.

[0064] Specifically, the dynamic sagittal plane construction module 3 includes a machine learning unit 301, an adaptive unit 302 and a historical data unit 303. The machine learning unit 301 is used to construct and optimize the human sagittal plane in real time based on the captured marker point data. The adaptive unit 302 is used to dynamically adjust the sagittal plane construction parameters according to the patient's body shape and posture changes. The historical data unit 303 is used to store and analyze historical construction data and optimize the sagittal plane construction algorithm.

[0065] The real-time feedback module 4 is used to generate a real-time feedback signal according to the distance between the marker point and the sagittal plane. The calculation method of the distance between the marker point and the sagittal plane is as follows (refer to Figure 8 ):

[0066] The equation of the plane perpendicular to line segment AB and with the midpoints of A(x1,y1,z1) and B(x2,y2,z2) is given as: 0, let the general formula of the plane equation be Ax+By+Cz+D=0 (which can be obtained by expanding and simplifying the previous equation), where A=x2-x1, B=y2-y1, C=z2-z1,

[0067] Suppose there is a point P(x0, y0, z0) in space. According to the distance formula from point (x0, y0, z0) to plane Ax+By+Cz+D=0

[0068] A=x2-x1,B=y2-y1,C=z2-z1, Substitute into the distance formula;

[0069] The calculation formula for the distance d from point P (x0, y0, z0) to the required sagittal plane is:

[0070] Specifically, the real-time feedback module 4 includes a dynamic adjustment unit 401, a multimodal feedback unit 402 and a threshold setting unit 403. The dynamic adjustment unit 401 is used to dynamically adjust the intensity and frequency of the feedback signal according to the progress of training. The multimodal feedback unit 402 is used to provide multiple feedback methods such as vision, hearing and touch to enhance the patient's perception. The threshold setting unit 403 is used to set the trigger threshold of the feedback signal according to the specific situation of the patient.

[0071] The display module 5 is used to display the captured marker point information, dynamic sagittal plane construction results and real-time feedback signals on the screen. Specifically, the display module 5 includes a visualization unit 501, a multi-view display unit 502 and a data analysis unit 503. The visualization unit 501 is used to display the distance between the marker point and the sagittal plane in a graphical manner in real time on the screen. The multi-view display unit 502 is used to display the positional relationship between the marker point and the sagittal plane from different perspectives. The data analysis unit 503 is used to display the training data in the form of a chart to facilitate patients and therapists to understand the progress of training.

[0072] In addition, the alarm module 6 is used to issue an alarm signal when the distance between the marking point and the sagittal plane exceeds a preset threshold, and it can also provide different levels of alarm signals according to the degree of distance exceeded. Specifically, the alarm signal is divided into three levels: mild (1mm-5mm), moderate (5mm-10mm), and severe (10mm). The alarm response time is ≤0.2s, and the alarm signal triggering formula is:

[0073]

[0074] Where A is the alarm signal level, and d is the distance between the marking point and the sagittal plane.

[0075] Example 2 (Comparison with traditional training methods)

[0076] This embodiment is based on the experiment of embodiment 1, and the specific experimental content is as follows:

[0077] Experimental subjects

[0078] A 14-year-old patient with AIS was tested and divided into an experimental group and a control group. The experimental group used the intelligent scoliosis rehabilitation training feedback system presented in this invention. The training goal was to bring the markings of the seventh cervical vertebra, the convex side of the thoracic spine, and the convex side of the lumbar spine as close to the sagittal plane as possible. The training period lasted for 6 weeks, with 30-minute training sessions three times per week. The control group used traditional physical therapy, including manual correction and conventional equipment training, for 6 weeks, with 30-minute training sessions three times per week.

[0079] Experimental process:

[0080] Experimental group:

[0081] Four industrial cameras are arranged in the training room to ensure that the field of view of each camera covers the patient's main training area.

[0082] Active markers were installed on the patient's body surface at the seventh cervical vertebra mark, the convex side of the thoracic vertebra, the convex side of the lumbar vertebra, and the midline of the sacrum.

[0083] Start the marker point spatial coordinate capture module and the dynamic sagittal plane construction module to capture the three-dimensional coordinates of the marker points in real time and construct the human body sagittal plane.

[0084] Through the display module, patients and therapists can view the distance between the marker point and the sagittal plane in real time and adjust the training posture according to the real-time feedback signal.

[0085] Control group:

[0086] Manual correction is performed by professional therapists and conventional equipment is used for training. There is no real-time feedback during the training process.

[0087] Experimental results:

[0088] Experimental group:

[0089] After 6 weeks of training, the patient's scoliosis angle decreased from the initial 25° to 18°, and the average distance between the marker point and the sagittal plane decreased from the initial 12mm to 6mm.

[0090] Control group:

[0091] After 6 weeks of training, the patient's scoliosis angle decreased from the initial 25° to 21°, and the average distance between the marker point and the sagittal plane decreased from the initial 12mm to 9mm.

[0092] Comparative conclusion: The training effect of the experimental group is significantly better than that of the control group. The real-time feedback and visualization functions of the system play a key role in improving the training effect.

[0093] Example 3 (Comparison of different training frequencies)

[0094] This embodiment is based on the experiment of embodiment 1, and the specific experimental content is as follows:

[0095] Experimental subjects

[0096] A 16-year-old AIS patient was tested and divided into a high-frequency group and a low-frequency group. The high-frequency group used the intelligent scoliosis rehabilitation training feedback system of the present invention. The training goal was to bring the markings of the seventh cervical vertebra, the convex side of the thoracic vertebra, and the convex side of the lumbar vertebra as close to the sagittal plane as possible. The training period was 8 weeks, with 45-minute training sessions four times per week. The low-frequency group used the intelligent scoliosis rehabilitation training feedback system of the present invention with the same training goal. The training period was 8 weeks, with 45-minute training sessions twice per week.

[0097] Experimental process:

[0098] Six industrial cameras are arranged in the training room to ensure that the field of view of each camera covers the patient's main training area.

[0099] Active markers were installed on the patient's body surface at the seventh cervical vertebra mark, the convex side of the thoracic vertebra, the convex side of the lumbar vertebra, and the midline of the sacrum.

[0100] Start the marker point spatial coordinate capture module and the dynamic sagittal plane construction module to capture the three-dimensional coordinates of the marker points in real time and construct the human body sagittal plane.

[0101] Through the display module, patients and therapists can view the distance between the marker point and the sagittal plane in real time and adjust the training posture according to the real-time feedback signal.

[0102] Experimental results:

[0103] High frequency group:

[0104] After 8 weeks of training, the patient's scoliosis angle decreased from the initial 30° to 22°, and the average distance between the marker point and the sagittal plane decreased from the initial 15mm to 8mm.

[0105] Low frequency group:

[0106] After 8 weeks of training, the patient's scoliosis angle decreased from the initial 30° to 25°, and the average distance between the marker point and the sagittal plane decreased from the initial 15mm to 11mm.

[0107] Comparative conclusion: The training effect of the high-frequency group was significantly better than that of the low-frequency group, indicating that increasing the training frequency is helpful to improve the rehabilitation effect.

[0108] Example 4 (Comparison of different system functions)

[0109] This embodiment is based on the experiment of embodiment 1, and the specific experimental content is as follows:

[0110] Subjects:

[0111] A 12-year-old patient with AIS was tested and divided into a full-function group and a simplified-function group. The full-function group used the intelligent scoliosis rehabilitation training and feedback system of the present invention, which includes a marker point spatial coordinate capture module, an active marker set module, a dynamic sagittal plane construction module, a real-time feedback module, and a display module. The training goal was to bring the marker points (the seventh cervical vertebra marker, the convex side of the thoracic vertebra, and the convex side of the lumbar vertebra) as close to the sagittal plane as possible. The training period lasted 12 weeks, with five 60-minute training sessions per week. The simplified-function group used a simplified version of the system, which only included the marker point spatial coordinate capture module and the display module. The training goal was the same, with five 60-minute training sessions per week for 12 weeks.

[0112] Experimental process:

[0113] Eight industrial cameras are arranged in the training room to ensure that the field of view of each camera covers the patient's main training area.

[0114] Active markers were installed on the patient's body surface at the seventh cervical vertebra mark, the convex side of the thoracic vertebra, the convex side of the lumbar vertebra, and the midline of the sacrum.

[0115] Start the marker point spatial coordinate capture module to capture the three-dimensional coordinates of the marker point in real time and display it through the display module.

[0116] Experimental results:

[0117] Full-featured group:

[0118] After 12 weeks of training, the patient's scoliosis angle decreased from the initial 20° to 12°, and the average distance between the marker point and the sagittal plane decreased from the initial 10mm to 4mm.

[0119] Simplified functional groups:

[0120] After 12 weeks of training, the patient's scoliosis angle decreased from the initial 20° to 16°, and the average distance between the marker point and the sagittal plane decreased from the initial 10mm to 7mm.

[0121] Comparative conclusion: The training effect of the full-function group was significantly better than that of the simplified-function group, indicating that the system's dynamic sagittal plane construction and real-time feedback functions played a key role in improving the training effect.

[0122] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. An intelligent scoliosis rehabilitation training feedback system, characterized by: include: A marker point spatial coordinate capture module (1), which is used to dynamically capture the three-dimensional coordinates of the patient's body surface marker points; Active marker group module (2), including but not limited to the seventh cervical vertebra marker, the thoracic vertebra convex side marker, the lumbar vertebra convex side marker, the left posterior superior iliac spine marker, and the right posterior superior iliac spine marker, which are marked at specific locations on the patient's body surface; Dynamic sagittal plane construction module (3) constructs the standard human anterior and posterior median sagittal plane based on the sacral midline as an important reference plane for rehabilitation training; A real-time feedback module (4) is used to generate a real-time feedback signal according to the distance between the calculated marker point and the constructed sagittal plane; A display module (5) is used to display the captured marker information, dynamic sagittal plane construction results and real-time feedback signals on a screen; The goal of the rehabilitation training is to bring the locations of specific marker points (the seventh cervical vertebra marker, the thoracic vertebra convex side marker, and the lumbar vertebra convex side marker) as close as possible to the sagittal plane. In addition, the marker point spatial coordinate capture module (1), the active marker group module (2), the dynamic sagittal plane construction module (3), the real-time feedback module (4), and the display module (5) are connected through network communication.

2. The intelligent scoliosis rehabilitation training feedback system according to claim 1, characterized in that: The marking point spatial coordinate capture module (1) includes multiple synchronized industrial cameras and adopts a high frame rate and high resolution configuration. The marking point spatial coordinate capture module (1) includes a synchronization unit (101), a calibration unit (102) and a multi-frame fusion unit (103). The synchronization unit (101) is used to ensure the synchronization of the multiple industrial cameras in time and space, and to capture the three-dimensional coordinates of the marking point with high precision. The calibration unit (102) is used to regularly perform spatial calibration on the cameras to ensure the capture accuracy. The multi-frame fusion unit (103) is used to fuse the data captured by the multiple cameras to improve the accuracy of the three-dimensional coordinates. The marking point spatial coordinate capture module (1) also includes a laser radar unit (104). The laser radar unit (104) and the industrial camera group constitute a multi-modal sensor array, and compensate for the blind spot of the optical camera's field of view through pulse time difference ranging.

3. The intelligent scoliosis rehabilitation training feedback system according to claim 2, characterized in that: The active marker group module (2) comprises a high contrast unit (201), a wireless communication unit (202), an adjustable unit (203) and a reflective material unit (204). The high contrast unit (201) is mainly used to enable the marker to maintain high contrast under different lighting conditions, so as to facilitate camera capture. The wireless communication unit (202) is used to transmit the status information of the marker to the system in real time. The adjustable unit (203) is used to adjust the position and size of the marker according to the patient's body shape to ensure the accuracy of the marking. The reflective material unit (204) is used to enhance the visibility of the marker.

4. The intelligent scoliosis rehabilitation training feedback system according to claim 1, characterized in that: The dynamic sagittal plane construction module (3) includes a machine learning unit (301), an adaptive unit (302) and a historical data unit (303). The machine learning unit (301) is used to construct and optimize the human sagittal plane in real time based on the captured marker point data. The adaptive unit (302) is used to dynamically adjust the construction parameters of the sagittal plane according to the patient's body shape and posture changes. The historical data unit (303) is used to store and analyze historical construction data and optimize the sagittal plane construction algorithm. The sagittal plane calculation method is as follows: Assume that there are two points in space: the left posterior superior iliac spine A (x1, y1, z1) and the right posterior superior iliac spine B (x2, y2, z2). First, find the coordinates of the midpoint M of AB: According to the midpoint coordinate formula, the coordinates of M are Then find the vector Because the dynamic sagittal plane passes through point M and is the normal vector (the plane is perpendicular to the line segment AB, then the direction vector of the line segment AB is the normal vector of the plane). According to the point normal equation of the plane A (xx m )+B(yy m )+C(zz m )=0,(where(x m ,y m , z m ) is a point M on the plane, (A, B, C) is the normal vector of the plane); Here: A=x2-x1, B=y2-y1, C=z2-z1; Then the equation of the plane is In addition, a specific marker point P(x0,y0.z 0) The calculation formula for the distance d to the required sagittal plane is:

5. The intelligent scoliosis rehabilitation training feedback system according to claim 1, characterized in that: The real-time feedback module (4) comprises a dynamic adjustment unit (401), a multimodal feedback unit (402) and a threshold setting unit (403). The dynamic adjustment unit (401) is used to dynamically adjust the intensity and frequency of the feedback signal according to the progress of training. The multimodal feedback unit (402) is used to provide multiple feedback modes such as vision, hearing and touch to enhance the patient's perception. The threshold setting unit (403) is used to set the trigger threshold of the feedback signal according to the specific situation of the patient.

6. The intelligent scoliosis rehabilitation training feedback system according to claim 1, characterized in that: The display module (5) comprises a visualization unit (501), a multi-view display unit (502) and a data analysis unit (503), wherein the visualization unit (501) is used to display the distance between a specific marking point and the sagittal plane in a graphical manner on the screen in real time. The multi-view display unit (502) is used to display the positional relationship between the marker points and the sagittal plane from different viewpoints, and the data analysis unit (503) is used to display the training data in a graphical form to facilitate patients and therapists to understand the training progress.

7. The intelligent scoliosis rehabilitation training feedback system according to claim 1, characterized in that: The system further comprises an alarm module (6), which is used to send out an alarm signal when the distance between the marking point and the sagittal plane exceeds a preset threshold, and can also provide alarm signals of different levels according to the degree of distance exceeded.

8. The intelligent scoliosis rehabilitation training feedback system according to claim 3, characterized in that: The synchronization error between the cameras is controlled within ±0.1ms, the calibration accuracy error is ≤0.5mm, the markers in the active marker group module (2) can maintain a contrast ratio of ≥80% within the range of light intensity from 50lux to 1000lux, the wireless communication delay is ≤10ms, and the marker position adjustment accuracy is ≤1mm.

9. The intelligent scoliosis rehabilitation training feedback system according to claim 1, characterized in that: The sagittal plane construction error in the dynamic sagittal plane construction module (3) needs to be ≤2mm, the training data set of the machine learning unit (301) contains ≥1000 patient data, and the adaptive unit (302) adjusts the response time to ≤0.5s.

10. The intelligent scoliosis rehabilitation training feedback system according to claim 5, characterized in that: The intensity range of the multimodal feedback unit (402) is 0% to 100%, the frequency range of the dynamic adjustment unit (401) is 0.1 Hz to 10 Hz, and the threshold setting unit (403) sets the threshold range to 1 mm to 20 mm.

Images