Postoperative rehabilitation posture intelligent monitoring system based on multi-modal data

Abstract

The application discloses a spine postoperative rehabilitation body state intelligent monitoring system based on multi-modal data, and particularly relates to the field of medical informatics and nursing guidance, and is used for solving the problem that in the existing spine postoperative rehabilitation nursing process, artificial observation is mainly relied on, and it is difficult to continuously monitor and predict risks of patient body state changes and potential compensatory behaviors; by fusing electronic medical record information, nursing records and body state motion data, an individualized dynamic activity safety domain is constructed, posture changes in each stage in the body position conversion process are matched and analyzed, abnormal body states and compensatory modes are identified, and a body position conversion stage correlation network is established to predict continuous compensatory risks, so that nursing risk prediction results and corresponding nursing intervention guidance strategies are generated, the safety of postoperative rehabilitation nursing is improved, an auxiliary decision basis is provided for nursing personnel, and intelligent nursing monitoring, precise risk assessment and individualized nursing intervention in the spine postoperative rehabilitation process are realized.

Description

Technical Field

[0001] This invention relates to the fields of medical informatics and nursing guidance technology, and more specifically, to an intelligent monitoring system for postoperative posture in spinal surgery rehabilitation based on multimodal data. Background Technology

[0002] Patients with spinal diseases who undergo vertebral fusion and internal fixation surgery typically require a long period of rehabilitation care. Postural changes, out-of-bed activities, and rehabilitation training are crucial components of the postoperative recovery phase. In nursing scenarios, patients need to coordinate multiple continuous movements, such as trunk lifting, lateral shift of the center of gravity, trunk forward leaning, lower limb exertion, and standing balance, during bed rest, turning over, sitting up, standing, and walking. However, due to the functional limitations of some spinal mobility units caused by the fusion segment, and the constraint of the internal fixation structure on the range of spinal motion, coupled with postoperative pain, nerve root irritation, decreased muscle strength, and weakened proprioception, patients are prone to compensatory movements such as excessive forward leaning of the trunk, lateral deviation, hand support, or pauses during postural changes or daily activities. These abnormal postures can not only increase the stress burden on the internal fixation structure but also increase the risk of falls, pressure sores, or worsening neurological symptoms, thus affecting the rehabilitation outcome.

[0003] Therefore, continuous monitoring of patients' postural characteristics, mobility, and functional recovery status, as well as timely identification of abnormal behaviors, are crucial during postoperative rehabilitation care. Current nursing methods largely rely on nurses' manual observation and experience, making it difficult to continuously and multidimensionally monitor changes in posture, turning movements, sitting and standing postures, and gait characteristics during bed rest, out-of-bed activities, and rehabilitation training. Furthermore, it is challenging to dynamically assess rehabilitation risks by comprehensively considering individual factors such as the distribution of surgical segments, the characteristics of internal fixation structures, bone healing progress, and historical nursing records. Traditional methods typically only make simple judgments on single changes in position, lacking analysis of the correlation between movements at different stages of position transitions, and are also insufficient for predicting potential continuous compensatory behaviors or nursing risks in advance.

[0004] Therefore, there is an urgent need for an intelligent monitoring system for postoperative rehabilitation posture based on multimodal data of spinal surgery. This system can dynamically monitor and identify abnormalities in the patient's postoperative posture by integrating multi-source physiological signals and kinematic data, and can be combined with the patient's surgical plan, rehabilitation stage and nursing plan. Summary of the Invention

[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide an intelligent post-operative posture monitoring system for spinal surgery rehabilitation based on multimodal data to address the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A multimodal data-based intelligent postural monitoring system for spinal surgery rehabilitation includes:

[0008] The basic data module is used to extract the number and distribution of surgical fusion segments, physical property parameters of internal fixation devices, and the location and orientation of disc herniation from the electronic medical record.

[0009] The range generation module is used to calculate the loss of the patient's overall spinal mobility unit based on the number and distribution of fused segments, calculate the allowable range of motion in the sagittal and coronal planes based on the physical characteristics of the internal fixation device, and combine the two with the bone healing histological stage corresponding to the postoperative nursing days to generate an individualized dynamic activity safety domain.

[0010] The nursing node annotation module is used to extract the completion time points of each sub-stage in the position change process from the nursing records, identify the auxiliary action records in the completion process of each sub-stage, and annotate the corresponding nursing attention nodes.

[0011] The matching module is used to match each nursing concern node with the individualized dynamic activity safety domain, and extract the extent and direction of the exceedance of nursing concern nodes that exceed the allowed activity range during the matching process.

[0012] The compensation identification module is used to perform three-dimensional spatial vector matching between the direction of the excess and the location of the intervertebral disc herniation, and to attribute the compensation pattern to the vector matching result by combining the muscle strength rating of the muscle groups involved in the force exertion. The compensation pattern includes pain avoidance compensation pattern, muscle weakness compensation pattern and proprioceptive impairment compensation pattern.

[0013] The risk assessment module is used to construct a network of relationships for each positional transition stage based on the positional transition process, mark the compensation patterns of each nursing concern node at the corresponding position in the network, calculate the propagation probability of the compensation pattern in subsequent stages, predict the set of risk nodes with continuous compensation, generate nursing risk prediction results, and generate nursing intervention guidance strategies.

[0014] As a further aspect of the present invention, in the basic data module, the distribution location of the fusion segments includes the vertebral segment number corresponding to each fusion segment and the adjacency relationship between each fusion segment; the physical property parameters of the internal fixation device include the material elastic modulus, structural configuration, cross-sectional moment of inertia, and effective working length.

[0015] As a further aspect of the present invention, the range generation module calculates the overall spinal mobility unit loss based on the number and distribution of fused segments, and calculates the allowable range of motion in the sagittal and coronal planes based on the physical characteristic parameters of the internal fixation device, specifically including:

[0016] Based on the anatomical reference values ​​of the sagittal extension range of motion, sagittal flexion range of motion, and coronal lateral flexion range of motion of the spinal functional units under normal physiological conditions, the anatomical reference values ​​of each fused segment in the corresponding direction are summed to obtain the range of motion loss of the entire spine in each direction.

[0017] Obtain the physical property parameters of the internal fixation device, establish the mechanical relationship between bending load and structural rotation angle, calculate the relationship between bending moment and rotation angle of the internal fixation device in the sagittal extension direction, sagittal flexion direction and coronal lateral flexion direction, and take the angle corresponding to the internal fixation device material yield strength as the initial allowable range of motion of the internal fixation device in each direction.

[0018] As a further aspect of the present invention, the generation of an individualized dynamic activity safety domain in the range generation module, which combines the bone healing histological stage corresponding to the number of postoperative nursing days, specifically includes:

[0019] Based on the number of postoperative nursing days, the histological stage of the current bone healing stage is estimated, the range of values ​​of the elastic modulus of the callus and the integration state of the bone-internal fixation interface are obtained for each histological stage, and the equivalent bending stiffness is calculated by considering the internal fixation and callus as a combined section according to the composite beam theory. The initial allowable range of motion is corrected by the equivalent bending stiffness to obtain the current allowable range of motion.

[0020] The current permissible range of motion, together with the overall spinal mobility loss value, constitutes an individualized dynamic activity safety domain consisting of the upper limit of sagittal extension, the upper limit of sagittal flexion, and the upper limit of coronal lateral flexion.

[0021] As a further aspect of the present invention, the nursing node annotation module identifies auxiliary action records during the completion of each sub-stage and annotates the corresponding nursing attention nodes, specifically including:

[0022] Extract the start and end times of the three sub-stages of body position change—lateral shift of center of gravity, forward tilt of trunk, and standing balance—from the nursing records, and calculate the actual time taken to complete each sub-stage.

[0023] Identify and record the auxiliary movements that occur during the completion of each sub-stage. The auxiliary movements include hand support, torso twisting, and pauses.

[0024] Sub-stages whose actual completion time exceeds the preset reference time range, as well as sub-stages with auxiliary action records, are collectively marked as nursing attention nodes.

[0025] As a further aspect of the present invention, the matching module specifically includes matching each nursing concern node with the individualized dynamic activity safety domain, including:

[0026] The corresponding anatomical motion plane is determined based on the positional transition sub-stage where the nursing attention node is located. The trunk forward tilt and standing balance sub-stages correspond to sagittal forward flexion and backward extension activities, respectively, and the center of gravity lateral shift stage corresponds to coronal lateral flexion activities. The upper limit of the activity corresponding to the anatomical motion plane where the nursing attention node is located is extracted from the individualized dynamic activity safety domain as the matching benchmark.

[0027] The actual trunk tilt angle reached during the positional change at each nursing attention point is obtained. The difference between the actual trunk tilt angle and the matching benchmark is calculated to obtain the excess range. The excess direction is determined based on the deflection direction of the actual trunk tilt angle relative to the longitudinal axis of the spine.

[0028] As a further aspect of the present invention, the compensation identification module performs three-dimensional spatial vector matching between the direction of the excess and the location of the intervertebral disc herniation, and combines the muscle strength rating of the muscle groups involved in the force exertion to attribute the compensation pattern to the vector matching results, specifically including:

[0029] Establish a three-dimensional spatial coordinate system with the center of the vertebral body of the herniated segment as the origin. Represent the orientation of the herniated disc and the direction of deviation as unit vectors pointing from the origin to the herniated side and the trunk deflection side, respectively. Calculate the spatial angle between the two vectors.

[0030] When the spatial angle is less than the preset angle threshold, it is determined that the direction of the excess is pointing to the side of the intervertebral disc herniation. The positional relationship between the intervertebral disc herniation segment and the nerve root recorded in the electronic medical record is retrieved. If there is a contact relationship between the intervertebral disc herniation segment and the nerve root, it is attributed to the pain avoidance compensation mode.

[0031] When the spatial angle is not less than the preset angle threshold or the relationship between the herniated disc segment and the nerve root does not meet the pain avoidance attribution criteria, the muscle groups involved in force control are determined according to the sub-stage of the nursing attention node.

[0032] The system queries the recent muscle strength rating of the muscle groups that exert force. When the muscle strength rating is lower than the set level, it is attributed to a compensatory mode of insufficient muscle strength. When the muscle strength rating is not lower than the set level, it is attributed to a compensatory mode of proprioceptive impairment.

[0033] As a further aspect of the present invention, the risk assessment module, which calculates the propagation probability of the compensation pattern in subsequent stages and predicts the set of risk nodes with continuous compensation, specifically includes:

[0034] In the sub-stages of the body position transition process, three sub-stages are added: trunk lifting, lower limb placement, and lower limb exertion. The corresponding abnormal state values ​​are calculated by the deviation of movement time and the degree of posture deviation.

[0035] A directed graph structure is constructed based on the preset execution order of the six sub-stages in the body position transition process: trunk lifting, center of gravity shifting laterally, lower limb placement, trunk leaning forward, lower limb exertion, and standing balance. Each sub-stage is treated as a node in the directed graph, and directed edges are established between adjacent stages according to the stage execution order.

[0036] The nursing attention nodes with marked compensation patterns are marked at the corresponding positions in the directed graph. The frequency of the occurrence of compensation patterns in each stage during the patient's previous multiple position changes is extracted from the nursing records. The conditional probability of the same type of compensation pattern appearing when pointing from the current node to the subsequent node along each directed edge is calculated as the propagation probability. The propagation probability is then corrected based on the abnormal state values ​​of adjacent sub-stage nodes.

[0037] Starting from the current nursing concern node, traverse subsequent nodes along the directed edge direction. Mark subsequent nodes whose propagation probability exceeds the threshold as continuous compensation risk nodes. Combine all continuous compensation risk nodes according to their appearance order in the directed graph to form a risk node set. Generate nursing risk prediction results based on the risk node set and generate corresponding nursing intervention guidance strategies.

[0038] The technical effects and advantages of the present invention: Intelligent Postural Monitoring System for Postural Rehabilitation after Spinal Surgery Based on Multimodal Data.

[0039] This invention integrates electronic medical records, nursing records, and patient postural movement data to perform multi-dimensional dynamic monitoring and analysis of postural transition behaviors during spinal surgery rehabilitation. It constructs an individualized dynamic activity safety domain for each patient, taking into full account individualized factors such as the distribution of surgical fusion segments, the characteristics of internal fixation structures, and the bone healing process, enabling quantitative assessment of postural changes. By identifying the characteristics of movements in each sub-stage of postural transitions and analyzing the extent and direction of abnormal postures, and combining this with attribution of compensatory patterns based on the location of herniated discs and muscle strength ratings, potential risk behaviors during rehabilitation can be accurately identified. Furthermore, by establishing a network linking postural transition stages, the propagation trend of compensatory patterns in subsequent movements can be predicted, thereby identifying continuous compensatory risk nodes in advance and generating nursing risk prediction results and nursing intervention guidance strategies. This helps nursing staff to take timely and targeted nursing measures, improving the accuracy and safety of postoperative rehabilitation nursing and effectively reducing the risk of complications. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the intelligent postural monitoring system for spinal surgery rehabilitation based on multimodal data, as described in this invention. Detailed Implementation

[0041] 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0042] Example 1

[0043] Figure 1 The present invention provides an intelligent postural monitoring system for spinal surgery rehabilitation based on multimodal data, comprising:

[0044] The basic data module is used to extract the number and distribution of surgical fusion segments, physical property parameters of internal fixation devices, and the location and orientation of disc herniation from the electronic medical record.

[0045] The range generation module is used to calculate the loss of the patient's overall spinal mobility unit based on the number and distribution of fused segments, calculate the allowable range of motion in the sagittal and coronal planes based on the physical characteristics of the internal fixation device, and combine the two with the bone healing histological stage corresponding to the postoperative nursing days to generate an individualized dynamic activity safety domain.

[0046] The nursing node annotation module is used to extract the completion time points of each sub-stage in the position change process from the nursing records, identify the auxiliary action records in the completion process of each sub-stage, and annotate the corresponding nursing attention nodes.

[0047] The matching module is used to match each nursing concern node with the individualized dynamic activity safety domain, and extract the extent and direction of the exceedance of nursing concern nodes that exceed the allowed activity range during the matching process.

[0048] The compensation identification module is used to perform three-dimensional spatial vector matching between the direction of the excess and the location of the intervertebral disc herniation, and to attribute the compensation pattern to the vector matching result by combining the muscle strength rating of the muscle groups involved in the force exertion. The compensation pattern includes pain avoidance compensation pattern, muscle weakness compensation pattern and proprioceptive impairment compensation pattern.

[0049] The risk assessment module is used to construct a network of relationships for each positional transition stage based on the positional transition process, mark the compensation patterns of each nursing concern node at the corresponding position in the network, calculate the propagation probability of the compensation pattern in subsequent stages, predict the set of risk nodes with continuous compensation, generate nursing risk prediction results, and generate nursing intervention guidance strategies.

[0050] In the basic data module, the distribution location of the fusion segments includes the vertebral segment number corresponding to each fusion segment and the adjacency relationship between each fusion segment; the physical property parameters of the internal fixation device include the material elastic modulus, structural configuration, cross-sectional moment of inertia and effective working length.

[0051] Surgical records are retrieved from the patient's electronic medical records and surgical notes. Each surgical segment in the record is analyzed item by item to identify the corresponding vertebral segment number. For example, when the fusion information is recorded as L3-L4, L4-L5, etc., it is parsed into the corresponding vertebral number and sorted according to the anatomical order of the spine to form a sequence of fusion segments. After identifying the vertebral segment numbers, the spatial arrangement relationship between the fusion segments is further determined. Specifically, adjacent vertebral segments are compared against the standard spinal segment anatomical sequence to determine whether there is a continuous relationship between adjacent fusion segments. For example, when the fusion segments are L3-L4 and L4-L5, they are determined to be continuous adjacent fusion structures; when the fusion segments are L2-L3 and L4-L5, they are determined to be non-continuous fusion structures. Subsequently, a fusion segment distribution information table is constructed based on the identified vertebral segment numbers and the adjacency relationships between fusion segments. This table records the starting and ending vertebral segment numbers of each fusion segment, as well as its spatial relationship with adjacent fusion segments, thus comprehensively characterizing the spatial distribution of fusion segments formed in the patient's spine postoperatively. This process, through standardized analysis of surgical segment information and combined with spinal anatomy for relationship determination, enables the spatial distribution of fusion segments to be clearly identified and used for subsequent assessment of spinal mobility.

[0052] The implanted internal fixation device is identified by reading the device model information recorded in the patient's surgical record. First, the corresponding material type is retrieved from a pre-established device parameter database based on the device model, and the elastic modulus parameter of the material is obtained from the database. For example, the elastic modulus of titanium alloy is approximately 110 GPa as a typical example value. Then, the structural configuration is determined based on the internal fixation structure used in the surgical record, such as a pedicle screw-connector rod structure or a plate fixation structure, and its stress characteristics are determined according to this structural form. After determining the structural configuration, the cross-sectional moment of inertia parameter is determined by reading the structural dimensions such as the diameter, thickness, or width of the internal fixation connecting rod or fixation plate, according to the standard cross-sectional moment of inertia calculation rules. For cylindrical connecting rod structures, the cross-sectional moment of inertia is determined by its diameter parameter; for plate structures, the corresponding cross-sectional moment of inertia is determined based on the thickness and width parameters. Simultaneously, the effective working length of the internal fixation device is determined based on its actual installation position between spinal segments. This length is obtained by measuring the center distance between adjacent fixation screws. For example, when the distance between adjacent pedicle screws is 45 mm, this distance is recorded as the effective working length of the corresponding connecting rod. Finally, the four types of parameters—material elastic modulus, structural configuration, moment of inertia, and effective working length—are integrated to form a set of physical property parameters for the internal fixation device, used to characterize the mechanical constraint characteristics of the internal fixation structure during spinal movement.

[0053] In the range generation module, the loss of the patient's overall spinal mobility unit is calculated based on the number and distribution of fused segments, and the allowable range of motion in the sagittal and coronal planes is calculated based on the physical property parameters of the internal fixation device.

[0054] A reference table for the range of motion of spinal functional units was established based on standard spinal anatomy data. This table records the typical range of motion of each vertebral segment under normal physiological conditions. For example, the typical range of motion of lumbar segments in sagittal flexion is approximately 12° to 18°, in sagittal extension is approximately 5° to 8°, and in coronal lateral flexion is approximately 4° to 7°. These ranges serve as examples of anatomical reference values ​​for range of motion. Subsequently, based on the distribution information of fused segments identified in the patient's surgical records, the spinal functional units involved in each fused segment were matched one by one. For example, when the fused segments are L3-L4 and L4-L5, the reference range of motion values ​​for the L3-L4 and L4-L5 functional units in sagittal flexion, sagittal extension, and coronal lateral flexion were extracted respectively. Next, the values ​​are accumulated in three anatomical movement directions: the forward flexion reference ranges of all fused segments are summed to obtain the sagittal forward flexion range of motion loss; the extension reference ranges of each fused segment are summed to obtain the sagittal extension range of motion loss; and the lateral flexion reference ranges of each fused segment are summed to obtain the coronal lateral flexion range of motion loss. For example, if a patient has two consecutive fused segments with forward flexion reference ranges of 15° and 14° respectively, the sum of these two ranges is 29°, which is taken as the patient's overall range of motion loss in the sagittal forward flexion direction. By statistically analyzing each fused segment and accumulating the values ​​in the three anatomical movement directions in the above manner, the overall range of motion loss of the patient's spine in the sagittal extension, sagittal forward flexion, and coronal lateral flexion directions is obtained, which characterizes the degree of decline in spinal mobility caused by segmental fusion.

[0055] The model of the internal fixation device was retrieved from the surgical record. Then, based on the installation method of the internal fixation structure in the spinal segment, its stress structure was determined. A correspondence between bending load and structural rotation angle was established according to the stress law of bending components in mechanics of materials. The bending deformation characteristics of the internal fixation structure under bending load were determined through specification calibration or experimental measurement, and the law of change in structural rotation angle as bending moment increases was determined. After establishing this mechanical relationship, the structural deformation of the internal fixation structure under bending load in the sagittal extension direction, sagittal flexion direction, and coronal lateral flexion direction were simulated. By gradually increasing the bending load and recording the resulting rotation angle changes, the correspondence curve between bending moment and structural rotation angle was obtained. When the structural rotation angle gradually increases until the internal stress of the material reaches the yield strength, it is considered that the rotation angle has reached the material's safe working limit. For example, when the titanium alloy connecting rod reaches a structural rotation angle of approximately 2° to 4° during bending, it is close to its material yield state. This corresponding angle is then taken as the maximum allowable rotation angle of the internal fixation structure in the corresponding direction. By determining the limit angle value in three anatomical movement directions, the initial permissible range of motion of the internal fixation device in the sagittal extension, sagittal flexion, and coronal lateral flexion directions is obtained, which describes the maximum angle range that the internal fixation structure can safely withstand during spinal movement.

[0056] In the range generation module, an individualized dynamic activity safety domain is generated by combining the bone healing histological stage corresponding to the number of postoperative nursing days.

[0057] The bone healing stage is estimated based on the number of postoperative care days. The nursing record clearly records the patient's surgery date and the number of daily postoperative care days. This number of care days is read and matched against a pre-established bone healing stage reference table to determine the current histological stage of bone healing. For example, in this embodiment, 1 to 2 weeks postoperatively corresponds to the inflammatory stage, 3 to 6 weeks postoperatively corresponds to the callus formation stage, 6 to 12 weeks postoperatively corresponds to the callus mineralization stage, and more than 3 months postoperatively corresponds to the bone remodeling stage. Subsequently, based on the determined histological stage, the range of callus elastic modulus values ​​for the corresponding stage is retrieved from the bone tissue biomechanical parameter table, and the integration state of the bone-internal fixation interface is simultaneously determined. For example, in the early stage, the interface is in a weakly connected state, while in the callus mineralization stage, the interface is in a stable connected state. After determining the above parameters, the internal fixation structure and callus tissue are considered as a combined structure sharing the bending load, and their overall bending resistance is evaluated based on composite beam theory. In the specific calculation process, the bending contribution of the internal fixation device is first determined based on its elastic modulus and moment of inertia. Then, the bending contribution of the callus structure is determined based on its elastic modulus and the cross-sectional dimensions of the area where the callus participates in the stress. The degree of callus participation in the stress is determined in conjunction with the integration state of the bone-internal fixation interface. When the interface integration state is stable, a higher value is taken for the proportion of callus participation in the stress, such as 70% as an example proportion. When the interface is in a weak connection state, a lower value is taken for the proportion of callus participation in the stress, such as 30% as an example proportion. The specific proportion is flexibly set according to the callus formation process to ensure that the sum of the proportions is 1. By comprehensively calculating the bending contributions of the internal fixation structure and the callus structure, the equivalent bending stiffness of the combined structure is obtained. Subsequently, this equivalent bending stiffness is used to modify the initial allowable range of motion obtained above, that is, the allowable rotation angle is expanded or maintained according to the degree of improvement in the overall bending capacity of the structure, thereby obtaining the patient's allowable range of motion at the current bone healing stage.

[0058] The current permissible range of motion is read, corresponding to the permissible structural rotation angles in the sagittal extension, sagittal flexion, and coronal lateral flexion directions. Simultaneously, the overall spinal mobility loss value, accumulated from the fused segments, is read, also corresponding to the three anatomical movement directions. The two datasets are then jointly analyzed, considering both the permissible structural range of motion and the mobility loss due to segment fusion in each anatomical movement direction. For example, if the permissible structural range of motion in the sagittal flexion direction is 20° and the mobility loss due to segment fusion is 10°, the maximum safe angle of motion achievable in that direction is the combined constraint value, recorded as approximately 10° as the upper limit of safe motion in this embodiment. The upper limits of safe motion in the sagittal extension and coronal lateral flexion directions are determined in the same manner. Finally, the upper limits of motion parameters in the three directions are combined to form an individualized dynamic safe range of motion. This safe range, with the maximum safe angle of motion in the three anatomical movement directions as its boundary, characterizes the range of spinal movements that the patient can safely complete under the current bone healing stage and internal fixation constraints, thereby achieving an individualized constraint description of the postoperative rehabilitation range of motion.

[0059] The nursing node annotation module identifies auxiliary action records during the completion of each sub-stage and annotates the corresponding nursing attention nodes.

[0060] The nursing records contain continuous movement logs of patients during positional training or out-of-bed activities. These records include the start and end times of each movement, as well as the names of the movements, as observed by nursing staff. The records are also synchronized with postural monitoring equipment located in the ward to record the patient's postural changes. When extracting information from each sub-stage, the movement process is identified according to predefined postural transition stage division rules. Specifically, the moment the patient's trunk begins to move to one side, accompanied by lateral pelvic shift, is marked as the start time of the lateral shift sub-stage. The moment the patient completes the lateral shift and reaches a stable sitting posture or prepares to lean forward is marked as the end time of the lateral shift sub-stage. Similarly, when the patient's trunk bends forward from an upright or semi-upright position and prepares to stand up, the moment this movement begins is marked as the start time of the trunk forward tilt sub-stage. The moment the trunk reaches its maximum forward tilt angle and enters the lower limb exertion stage is marked as the end time of the trunk forward tilt sub-stage. Finally, when the patient completes the standing movement and enters an upright posture, the moment the body leaves the bed or chair support is marked as the start time of the standing balance sub-stage. The moment the patient maintains a stable standing posture for a predetermined observation time is marked as the end time of the standing balance sub-stage. After extracting the start and end times of each sub-stage, the actual completion time of each sub-stage is obtained by calculating the time difference between the corresponding time points.

[0061] The assistive movement type is identified by motion capture of the patient's postural changes. Similar to the patient positional transition capture method, motion capture is accomplished through a combination of visual acquisition devices installed at the bedside in the ward or rehabilitation training area and wearable posture sensors. The visual acquisition devices are used to acquire information on postural changes of key parts of the patient's body, such as changes in the position of the shoulders, hips, and trunk center. The wearable posture sensors are used to record trunk tilt angles and body rotation changes. By continuously analyzing the acquired postural data, when the patient touches the bed surface, armrests, or other supports with both hands or one hand during a positional transition, it is identified as a hand support movement; when the trunk rotates significantly relative to the body's central axis with a rotation angle exceeding approximately 15° (as an example threshold), it is identified as a trunk twisting movement; when the patient maintains a static posture for more than approximately 2 seconds in a certain sub-stage (as an example threshold), it is identified as a pause or waiting movement. These assistive movements are recorded as the assistive movement type for the corresponding sub-stage after identification. Simultaneously, reference time ranges for each positional transition stage are established based on nursing experience data. For example, the reference time for the lateral shift of the center of gravity stage is 1 to 3 seconds, the reference time for the forward tilt of the trunk stage is 1 to 2.5 seconds, and the reference time for the standing balance stage is 2 to 4 seconds. When the actual completion time exceeds the upper limit of the corresponding reference range, the stage is judged to be abnormal. Finally, sub-stages that meet any of the following conditions are marked as nursing concern nodes: first, the actual completion time of the sub-stage exceeds the reference time range; second, the sub-stage identifies auxiliary movements such as hand support, trunk twisting, or pauses and waiting, thereby completing the identification and marking of nursing concern nodes.

[0062] The matching module specifically includes matching each nursing concern node with the individualized dynamic activity safety domain, including:

[0063] The corresponding anatomical motion plane is determined based on the positional transition sub-stage where the nursing attention node is located. The trunk forward tilt and standing balance sub-stages correspond to sagittal forward flexion and backward extension activities, respectively, and the center of gravity lateral shift stage corresponds to coronal lateral flexion activities. The upper limit of the activity corresponding to the anatomical motion plane where the nursing attention node is located is extracted from the individualized dynamic activity safety domain as the matching benchmark.

[0064] The actual trunk tilt angle reached during the positional change at each nursing attention point is obtained. The difference between the actual trunk tilt angle and the matching benchmark is calculated to obtain the excess range. The excess direction is determined based on the deflection direction of the actual trunk tilt angle relative to the longitudinal axis of the spine.

[0065] In the compensation identification module, the direction of the excess is matched with the location of the intervertebral disc herniation in three-dimensional space. The compensation pattern is attributed to the vector matching result by combining the muscle strength rating of the muscle groups involved in the force exertion.

[0066] Information on the herniated disc segment and its location is obtained from the patient's electronic medical record and imaging examination records. For example, the imaging report may state "L4-L5 disc herniation on the right posterolateral side." Then, based on the vertebral body structure corresponding to that segment, the geometric center of the vertebral body is determined as the origin of a three-dimensional spatial coordinate system. Spatial coordinate axes are established according to standard anatomical orientations, with the anterior-posterior direction as the sagittal axis, the lateral direction as the coronal axis, and the superior-inferior direction as the longitudinal axis. After establishing the coordinate system, the herniation direction is determined based on the location recorded in the imaging report. Specifically, when the herniation is on the right posterolateral side, a direction vector pointing from the vertebral body center to the right posterolateral side is determined in the three-dimensional coordinate system, and this direction vector is normalized to form a unit vector representing the location of the herniated disc. Simultaneously, the direction of deviation is determined based on the patient's trunk tilt direction obtained during posture monitoring. For example, when the patient's trunk tilts to the right during a positional change, a direction vector pointing from the vertebral body center to the side of trunk tilt is determined in the same coordinate system, and this vector is also normalized to form a unit vector. After establishing two directional vectors, the spatial angle between them is calculated by comparing their directional relationship in three-dimensional space. This reflects the degree of spatial consistency between the patient's trunk deflection direction and the direction of the intervertebral disc herniation. For example, when the angle between the two vectors is close to 0°, it indicates that their directions are basically consistent, while when the angle is close to 90°, it indicates a significant difference in direction.

[0067] A threshold determination is performed on the spatial angle calculated above. This threshold is set based on clinical experience; in this example, 30° is selected as the example threshold, meaning that when the spatial angle is less than 30°, the trunk deflection direction is considered to be significantly consistent with the direction of the intervertebral disc herniation. After this determination, the patient's imaging report or preoperative examination record is retrieved from the electronic medical record, and the description of the spatial relationship between the herniated segment and the nerve root is read. For example, the report records "the herniated disc is compressing or contacting the right nerve root." When the record shows that the herniated disc is in contact or compressing the nerve root, it is determined that the herniation may cause nerve stimulation or pain response. In this case, when the patient exhibits trunk deflection behavior consistent with the direction of the herniation during body position changes, it is considered that this behavior is associated with pain stimulation, and this abnormal posture is attributed to a pain avoidance compensatory pattern. This pattern represents the patient changing trunk posture during body position changes to reduce the stimulation of the nerve root by the herniated segment, thereby forming a compensatory postural change in a specific direction. When the spatial angle is greater than or equal to 30° or no contact relationship between the herniated disc and nerve root is found in the imaging record, the current trunk lateralization behavior is considered not to be pain avoidance compensation. Subsequently, based on the positional transition sub-stage at which the nursing focus point is located, the main muscle groups involved in force exertion at that stage are determined. For example, in the trunk forward tilting stage, the rectus abdominis and iliopsoas muscles are mainly involved; in the lateral shift of the center of gravity stage, the quadratus lumborum and oblique abdominal muscles are mainly involved; and in the standing balance stage, the erector spinae and gluteus medius muscles are mainly involved. After determining the corresponding muscle group, the most recent muscle strength rating result for that muscle group is retrieved from the electronic medical record or rehabilitation assessment record. The muscle strength rating adopts the clinically commonly used muscle strength grading standard, with a grade range of 0 to 5. In this embodiment, level three is used as the threshold for judging muscle strength function. When the muscle strength rating is lower than the set level, it means that the muscle group cannot complete stable force exertion under anti-gravity conditions. At this time, the compensatory behavior is judged as a compensatory mode of insufficient muscle strength. When the muscle strength rating reaches or exceeds level three, it means that the muscle group has basic anti-gravity ability, but abnormal posture still occurs. Then, it is determined that the abnormal posture is related to the decline in proprioceptive control ability and is attributed to a compensatory mode of proprioceptive impairment.

[0068] The risk assessment module calculates the propagation probability of the compensation pattern in subsequent stages and predicts the set of risk nodes where continuous compensation occurs.

[0069] According to the nursing training process, the continuous movements of a patient transitioning from a supine to a standing position are divided into six stages: trunk lifting stage, center of gravity shifting stage, lower limb positioning stage, trunk forward tilting stage, lower limb exertion stage, and standing balance stage. Specifically, the trunk lifting stage is defined as the movement of the patient's upper back leaving the bed or chair support; the center of gravity shifting stage is defined as the movement of shifting the body's center of gravity to one side and completing posture adjustment; the lower limb positioning stage is defined as the movement of the patient moving both lower limbs to a stable position suitable for standing; the trunk forward tilting stage is defined as the movement of the patient bending the trunk forward to prepare for standing; the lower limb exertion stage is defined as the movement of the patient using the lower limb muscle groups to lift the body off the chair or bed; and the standing balance stage is defined as the movement of the patient maintaining a stable upright posture after standing. After completing the phase division, the six phases are arranged in chronological order according to the fixed sequence of actions performed in nursing training. That is, after the trunk lifting phase, the center of gravity shifting phase begins; after the center of gravity shifting phase, the lower limb positioning phase begins; after the lower limb positioning phase, the trunk forward tilting phase begins; after the trunk forward tilting phase, the lower limb exertion phase begins; and after the lower limb exertion phase, the standing balance phase begins.

[0070] The three newly added sub-stages extract information on action duration and posture changes. Action duration is obtained by reading the time difference between the start and end times of the stage, while posture deviation is obtained by acquiring the offset angle or distance of the patient's trunk center point or pelvic center point relative to the body's longitudinal axis using a posture monitoring device. To determine if there are any abnormalities in the action execution, the actual action duration of each stage is compared with the reference time range recorded in the nursing guidelines. For example, in this embodiment, the reference time range for the trunk lifting stage is set to 1 to 2 seconds, for the lower limb placement stage to 1 to 2.5 seconds, and for the lower limb exertion stage to 1 to 3 seconds. When the actual duration exceeds the corresponding range, a time deviation is determined. At the same time, the posture deviation value is compared with the standard posture range. For example, when the trunk deviation angle exceeds approximately 10°, a posture deviation is determined. Subsequently, the action duration deviation value and posture deviation degree are uniformly scaled to convert the two types of indicators into numerical ranges with the same dimensions. In this embodiment, they are converted to values ​​between 0 and 1, and the two indicators are comprehensively calculated to obtain the abnormal state value of the stage.

[0071] Subsequently, each stage is used as a node in the graph structure, and directional connections are established between adjacent stages in the above order. Each node is only connected to the nodes of its preceding and following stages, thus forming a directed graph structure that reflects the sequence of body position transitions. This directed graph structure clearly shows the sequential relationships between each action stage and the possible action propagation paths, providing a structured representation for subsequent analysis of compensatory pattern propagation.

[0072] The nursing attention nodes that have been identified and whose compensatory patterns have been attributed are marked in the constructed six-stage directed graph structure. Then, data from the patient's previous postural change training is retrieved from the nursing records. The occurrence of the same type of compensatory pattern in each stage of each training session is statistically analyzed, and the number of times the same type of compensatory pattern appears in the next stage (which must be the sub-stage corresponding to the nursing attention node) after appearing in a certain stage is recorded. For example, the number of times the pain avoidance compensatory pattern appears in the trunk forward tilt stage after appearing in the center of gravity shift stage is counted, along with the total number of times the compensatory pattern appears in the center of gravity shift stage. By accumulating statistics from past records, the frequency relationship of compensatory pattern transitions between different stages is obtained, and the conditional probability of the same type of compensatory pattern appearing between adjacent stages is calculated. This probability value is recorded as the basic propagation probability on the corresponding directed edge. Then, the abnormal state values ​​of the nodes adjacent to this directed edge are read, and the basic propagation probability is corrected based on the abnormal state values. This is done by dividing the abnormal state values ​​into different value ranges and setting corresponding propagation probability correction amounts for different ranges. For example, when the abnormal state value is greater than 0.6, the propagation probability is increased by approximately 20% based on the base propagation probability; when the abnormal state value is between 0.3 and 0.6, the base propagation probability remains unchanged; and when the abnormal state value is less than 0.3, the propagation probability is decreased by approximately 10% based on the base propagation probability. The base propagation probability is adjusted according to the range in which the abnormal state value falls.

[0073] Starting with the currently detected nursing concern node, the system visits subsequent nodes one by one in the directed graph structure according to the action execution sequence. For example, when the current nursing concern node is in the trunk forward leaning stage, the standing balance node is visited sequentially according to the stage order. During the traversal, the final propagation probability corresponding to each directed edge is read and compared with a pre-set risk judgment threshold. This threshold is set based on the statistical results of historical nursing data; that is, when the propagation probability is greater than this value, it is determined that there is a high probability that the same compensatory behavior will continue in the subsequent stage. When the propagation probability of a subsequent node exceeds the threshold, the node is marked as a continuous compensation risk node, and its corresponding body position transition stage name and compensation mode type are recorded. Subsequently, all marked risk nodes are sorted according to the order of body position transition stages in the directed graph and combined to form a risk node set, which includes the non-nursing concern nodes passed through. For example, if continuous compensation risk is identified in both the trunk forward leaning and standing balance stages in a certain analysis, these two stages and the lower limb exertion in between are combined into a risk node set. Finally, nursing risk prediction results are generated based on the set of risk nodes. These results clearly indicate the types of abnormal postures that patients may continue to exhibit during subsequent positional transitions. Corresponding nursing intervention guidance strategies are generated based on different compensation modes. For example, when continuous compensation due to muscle weakness is predicted, nursing staff are advised to increase lower limb support or adjust the standing rhythm, thereby completing the process of nursing risk prediction and nursing intervention guidance generation.

[0074] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0075] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0076] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0077] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0078] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0079] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0080] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0081] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0082] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A postural intelligent monitoring system for spinal surgery rehabilitation based on multimodal data, characterized in that, include: The basic data module is used to extract the number and distribution of surgical fusion segments, physical property parameters of internal fixation devices, and the location and orientation of disc herniation from the electronic medical record. The range generation module is used to calculate the loss of the patient's overall spinal mobility unit based on the number and distribution of fused segments, calculate the allowable range of motion in the sagittal and coronal planes based on the physical characteristics of the internal fixation device, and combine the two with the bone healing histological stage corresponding to the postoperative nursing days to generate an individualized dynamic activity safety domain. The nursing node annotation module is used to extract the completion time points of each sub-stage in the position change process from the nursing records, identify the auxiliary action records in the completion process of each sub-stage, and annotate the corresponding nursing attention nodes. The matching module is used to match each nursing concern node with the individualized dynamic activity safety domain, and extract the extent and direction of the exceedance of nursing concern nodes that exceed the allowed activity range during the matching process. The compensation identification module is used to perform three-dimensional spatial vector matching between the direction of the excess and the location of the intervertebral disc herniation, and to attribute the compensation pattern to the vector matching result by combining the muscle strength rating of the muscle groups involved in the force exertion. The compensation pattern includes pain avoidance compensation pattern, muscle weakness compensation pattern and proprioceptive impairment compensation pattern. The risk assessment module is used to construct a network of relationships for each positional transition stage based on the positional transition process, mark the compensation patterns of each nursing concern node in the corresponding position in the network, calculate the propagation probability of the compensation pattern in the subsequent stage, predict the set of risk nodes with continuous compensation, generate nursing risk prediction results and generate nursing intervention guidance strategies. The range generation module, which combines the postoperative nursing days with the bone healing histological stage to generate an individualized dynamic activity safety domain, specifically includes: Based on the number of postoperative nursing days, the histological stage of the current bone healing stage is estimated, the range of values ​​of the elastic modulus of the callus and the integration state of the bone-internal fixation interface are obtained for each histological stage, and the equivalent bending stiffness is calculated by considering the internal fixation and callus as a combined section according to the composite beam theory. The initial allowable range of motion is corrected by the equivalent bending stiffness to obtain the current allowable range of motion. The current allowed range of motion, together with the overall spinal mobility loss value, constitutes an individualized dynamic activity safety domain consisting of the upper limit of sagittal extension, the upper limit of sagittal flexion, and the upper limit of coronal lateral flexion. In the compensation identification module, the direction of the excess is matched with the location of the herniated disc in three-dimensional space. The compensation pattern is attributed to the vector matching result by combining the muscle strength rating of the muscle groups involved in the force exertion. Specifically, this includes: Establish a three-dimensional spatial coordinate system with the center of the vertebral body of the herniated segment as the origin. Represent the orientation of the herniated disc and the direction of deviation as unit vectors pointing from the origin to the herniated side and the trunk deflection side, respectively. Calculate the spatial angle between the two vectors. When the spatial angle is less than the preset angle threshold, it is determined that the direction of the excess is pointing to the side of the intervertebral disc herniation. The positional relationship between the intervertebral disc herniation segment and the nerve root recorded in the electronic medical record is retrieved. If there is a contact relationship between the intervertebral disc herniation segment and the nerve root, it is attributed to the pain avoidance compensation mode. When the spatial angle is not less than the preset angle threshold or the relationship between the herniated disc segment and the nerve root does not meet the pain avoidance attribution criteria, the muscle groups involved in force control are determined according to the sub-stage of the nursing attention node. The system queries the recent muscle strength rating of the muscle groups that exert force. When the muscle strength rating is lower than the set level, it is attributed to a compensatory mode of insufficient muscle strength. When the muscle strength rating is not lower than the set level, it is attributed to a compensatory mode of proprioceptive impairment. The risk assessment module calculates the propagation probability of the compensation pattern in subsequent stages and predicts the set of risk nodes with continuous compensation, specifically including: In the sub-stages of the body position transition process, three sub-stages are added: trunk lifting, lower limb placement, and lower limb exertion. The corresponding abnormal state values ​​are calculated by the deviation of movement time and the degree of posture deviation. A directed graph structure is constructed based on the preset execution order of the six sub-stages in the body position transition process: trunk lifting, center of gravity shifting laterally, lower limb placement, trunk leaning forward, lower limb exertion, and standing balance. Each sub-stage is treated as a node in the directed graph, and directed edges are established between adjacent stages according to the stage execution order. The nursing attention nodes with marked compensation patterns are marked at the corresponding positions in the directed graph. The frequency of the occurrence of compensation patterns in each stage during the patient's previous multiple position changes is extracted from the nursing records. The conditional probability of the same type of compensation pattern appearing when pointing from the current node to the subsequent node along each directed edge is calculated as the propagation probability. The propagation probability is then corrected based on the abnormal state values ​​of adjacent sub-stage nodes. Starting from the current nursing concern node, traverse subsequent nodes along the directed edge direction. Mark subsequent nodes whose propagation probability exceeds the threshold as continuous compensation risk nodes. Combine all continuous compensation risk nodes according to their appearance order in the directed graph to form a risk node set. Generate nursing risk prediction results based on the risk node set and generate corresponding nursing intervention guidance strategies.

2. The intelligent postural monitoring system for spinal surgery rehabilitation based on multimodal data according to claim 1, characterized in that, In the basic data module, the distribution location of the fusion segments includes the vertebral segment number corresponding to each fusion segment and the adjacency relationship between each fusion segment; the physical property parameters of the internal fixation device include the material elastic modulus, structural configuration, cross-sectional moment of inertia and effective working length.

3. The intelligent postural monitoring system for spinal surgery rehabilitation based on multimodal data according to claim 1, characterized in that, In the range generation module, the loss of the patient's overall spinal mobility unit is calculated based on the number and distribution of fused segments, and the allowable range of motion in the sagittal and coronal planes is calculated based on the physical characteristics of the internal fixation device. Specifically, this includes: Based on the anatomical reference values ​​of the sagittal extension range of motion, sagittal flexion range of motion, and coronal lateral flexion range of motion of the spinal functional units under normal physiological conditions, the anatomical reference values ​​of each fused segment in the corresponding direction are summed to obtain the range of motion loss of the entire spine in each direction. Obtain the physical property parameters of the internal fixation device, establish the mechanical relationship between bending load and structural rotation angle, calculate the relationship between bending moment and rotation angle of the internal fixation device in the sagittal extension direction, sagittal flexion direction and coronal lateral flexion direction, and take the angle corresponding to the internal fixation device material yield strength as the initial allowable range of motion of the internal fixation device in each direction.

4. The intelligent postural monitoring system for spinal surgery rehabilitation based on multimodal data according to claim 1, characterized in that, The nursing node annotation module identifies auxiliary action records during the completion of each sub-stage and annotates the corresponding nursing attention nodes, specifically including: Extract the start and end times of the three sub-stages of body position change—lateral shift of center of gravity, forward tilt of trunk, and standing balance—from the nursing records, and calculate the actual time taken to complete each sub-stage. Identify and record the auxiliary movements that occur during the completion of each sub-stage. The auxiliary movements include hand support, torso twisting, and pauses. Sub-stages whose actual completion time exceeds the preset reference time range, as well as sub-stages with auxiliary action records, are collectively marked as nursing attention nodes.

5. The intelligent postural monitoring system for spinal surgery rehabilitation based on multimodal data according to claim 1, characterized in that, The matching module specifically includes matching each nursing concern node with the individualized dynamic activity safety domain, including: The corresponding anatomical motion plane is determined based on the positional transition sub-stage where the nursing attention node is located. The trunk forward tilt and standing balance sub-stages correspond to sagittal forward flexion and backward extension activities, respectively, and the center of gravity lateral shift stage corresponds to coronal lateral flexion activities. The upper limit of the activity corresponding to the anatomical motion plane where the nursing attention node is located is extracted from the individualized dynamic activity safety domain as the matching benchmark. The actual trunk tilt angle reached during the positional change at each nursing attention point is obtained. The difference between the actual trunk tilt angle and the matching benchmark is calculated to obtain the excess range. The excess direction is determined based on the deflection direction of the actual trunk tilt angle relative to the longitudinal axis of the spine.

Images