Method and system for generating an early intervention program for scoliosis based on multidisciplinary collaboration

Abstract

This invention discloses a method and system for generating early intervention plans for scoliosis based on multidisciplinary collaboration, belonging to the field of medical information technology. The method includes: collecting multi-source structured patient characteristic data and standardizing it to generate a standardized patient characteristic dataset; performing multi-dimensional assessments on the standardized patient characteristic dataset to generate a multidisciplinary assessment result set; based on the multidisciplinary assessment result set, obtaining the current stage of scoliosis development of the target patient; assigning a decision-making role to each discipline at the current stage according to a predefined development stage discipline role mapping relationship, forming a discipline role allocation structure; and generating professional intervention suggestions for the corresponding disciplines based on the discipline role allocation structure and the multidisciplinary assessment result set, summarizing them to form a multidisciplinary intervention suggestion set. This invention achieves automatic arbitration of topological constraints, dynamic role definition, and accurate generation of personalized executable plans.

Description

Technical Field

[0001] This invention relates to the field of medical information technology, and in particular to a method and system for generating early intervention programs for scoliosis based on multidisciplinary collaboration. Background Technology

[0002] Early intervention for scoliosis involves collaboration among multiple disciplines, including orthopedics, rehabilitation, and sports medicine. Existing technologies often use static staging standards to guide clinical decision-making and frequently use fixed weights or linear rules to simply overlay and integrate recommendations from various disciplines. While such methods achieve basic protocol generation, their decision-making logic often appears rigid and difficult to adapt to individual differences when faced with complex changes in growth potential.

[0003] The shortcomings of existing technologies lie in the lack of a dynamic definition mechanism for the decision-making roles of different disciplines. In the process of generating intervention plans, each discipline is usually integrated on an equal footing, failing to take into account the essential nonlinear differences in the clinical decision-making contributions of each discipline at different stages of scoliosis development. When suggestions from different disciplines conflict, there is a lack of automatic arbitration logic based on topological constraints, and manual negotiation or simple truncation is usually relied upon, making it difficult for the plan to balance safety and effectiveness, and unable to adaptively adjust the decision boundaries according to changes in the growth stage. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a method for generating early intervention plans for scoliosis based on multidisciplinary collaboration, which solves the problem in the prior art that the lack of dynamic disciplinary role definition and topological constraint arbitration logic makes it difficult to balance safety and effectiveness in plan generation.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for generating an early intervention program for scoliosis based on multidisciplinary collaboration, which includes collecting multi-source structured patient feature data and performing standardized processing to generate a standardized patient feature dataset. Perform multidimensional assessments on standardized patient characteristic datasets to generate a multidisciplinary assessment result set; Based on the multidisciplinary assessment results set, the current stage of scoliosis development of the target patient is obtained. According to the predefined development stage discipline role mapping relationship, each discipline is assigned a decision-making role under the current stage, forming a discipline role allocation structure. Based on the disciplinary role allocation structure and the multidisciplinary assessment results set, professional intervention suggestions for the corresponding disciplines are generated and summarized to form a multidisciplinary intervention suggestion set; Based on a multidisciplinary set of intervention suggestions and a disciplinary role allocation structure, and in accordance with predefined disciplinary decision boundary rules, conflict types among different disciplinary intervention suggestions are identified. According to role priority rules, conflicting suggestions are collaboratively and iteratively adjusted to generate a multidisciplinary collaborative consensus scheme. By leveraging the individual environmental adaptation characteristics of target patients, we can screen the feasibility and intensity of multidisciplinary collaborative consensus programs and generate structured individualized early intervention programs.

[0007] As a preferred embodiment of the method for generating early intervention programs for scoliosis based on multidisciplinary collaboration as described in this invention, the standardized patient feature dataset includes: Collect age, gender, bone age, spinal imaging, body shape, motor behavior, and daily living behavior information of the target patients to generate multi-source structured patient feature data. Perform missing value imputation and dimensional unification processing to generate multi-source structured patient feature data to be standardized. Standardized multi-source structured patient feature data is processed by format conversion and semantic encoding according to unified clinical data standards to generate a standardized patient feature dataset.

[0008] As a preferred embodiment of the method for generating early intervention plans for scoliosis based on multidisciplinary collaboration as described in this invention, the multidisciplinary assessment result set includes: Based on a standardized patient characteristic dataset, preliminary specialist assessment results are generated by calling orthopedic assessment rules, rehabilitation medicine assessment rules, sports medicine assessment rules, growth and development assessment rules, and psychological and behavioral assessment rules respectively. The preliminary specialty assessment results are encoded according to unified semantic tags and classification coding rules to generate coded specialty assessment results, and then aggregated to generate a multidisciplinary assessment result set.

[0009] As a preferred embodiment of the method for generating early intervention plans for scoliosis based on multidisciplinary collaboration as described in this invention, the subject role allocation structure includes: Based on the multidisciplinary assessment results set, the Cobb angle difference, bone age grading label and secondary sexual characteristic development label at continuous time points are extracted and input into the growth potential partitioning logic. By comparing the joint distribution interval of bone age and Risser sign, the growth active area, slowing area and stable area are divided to obtain the current stage of scoliosis development of the target patient. To determine the current stage of scoliosis development in the target patient, we traversed the time-series records of multidisciplinary joint interventions under different progression trajectories in the historical scoliosis case database and constructed an undirected weighted network with disciplines as nodes and the correlation of intervention effectiveness as edges. An improved spectral clustering algorithm is used to segment the network. The improved spectral clustering algorithm introduces stage labels as constraints after the Laplacian matrix eigenvalue decomposition to force the differentiation of subject clustering centers at different growth stages, generate subject role prototypes at each stage, and form an asymmetric mapping spectrum. The high-risk compensation pattern labels in the multidisciplinary assessment result set are input into the path matching interface of the asymmetric mapping graph. The initial role combination is resolving conflicts and calibrating boundaries through topological sorting within the graph to obtain the decision-making role of each discipline at the current stage. The decision-making roles of various disciplines at the current stage are summarized and indexed to establish an association between the target patient's current stage of scoliosis development, thus forming a structure for assigning disciplinary roles.

[0010] As a preferred embodiment of the method for generating early intervention plans for scoliosis based on multidisciplinary collaboration as described in this invention, the multidisciplinary intervention suggestion set includes: Based on the decision-making roles and authority boundaries defined in the subject role allocation structure, and combined with the quantitative indicators in the multidisciplinary evaluation result set, professional intervention suggestions for the corresponding subjects are generated. The professional intervention suggestions for the corresponding disciplines are structured and encapsulated according to unified clinical semantic coding rules to generate encapsulated discipline intervention suggestions. These suggestions are then aggregated and indexed to generate a multidisciplinary intervention suggestion set.

[0011] As a preferred embodiment of the method for generating early intervention plans for scoliosis based on multidisciplinary collaboration as described in this invention, the multidisciplinary collaborative consensus plan includes: Based on a multidisciplinary set of intervention recommendations and a disciplinary role allocation structure, a predefined disciplinary decision boundary rule is generated by constructing an interdisciplinary taboo graph. The interdisciplinary taboo graph uses a hypergraph structure to characterize the asymmetric exclusion relationship between the absolute taboo domain of orthopedics and the relatively safe domain of rehabilitation medicine. Based on the topological distance of the exclusion relationship, hard boundaries and flexible boundaries are divided to establish the overlapping operational range of different disciplinary recommendations. The action amplitude, execution frequency, and contraindication range from the multidisciplinary intervention suggestion set are input into a conflict detection engine based on an interdisciplinary contraindication map. By calculating the projection position of the parameter vectors of action amplitude, execution frequency, and contraindication range in the map manifold space, the conflict type currently existing in the target patient is identified. The conflict type includes hard boundary crossing conflict and flexible boundary overlapping conflict. The conflict type and subject role allocation structure are input into the collaborative iterative adjustment logic based on role priority. The collaborative iterative adjustment logic adopts the coordinate gradient descent method with mask. During the iteration process, the suggestion parameter coordinates corresponding to the dominant role are frozen. The overlapping parameters of the supporting role and the constraint role are fine-tuned along the negative gradient direction of the graph and returned to the overlapping operation range to generate the adjusted multidisciplinary intervention suggestion set. The adjusted set of multidisciplinary intervention suggestions is then input into the interdisciplinary taboo graph for boundary re-verification. Conflict detection and collaborative iterative adjustment are performed repeatedly until there are no hard boundary crossing conflicts, generating a multidisciplinary collaborative consensus scheme.

[0012] As a preferred embodiment of the method for generating early intervention programs for scoliosis based on multidisciplinary collaboration as described in this invention, the structured individual early intervention program includes: By utilizing the individual environmental adaptation characteristics of target patients, the family executive capacity matching degree of intervention measures in the multidisciplinary collaborative consensus plan is tested, and a set of intervention measures after feasibility filtering is generated. The set of intervention measures after feasibility filtering and the growth potential information of the target patients are input into the intensity grading logic. The intervention intensity is discretized and calibrated according to the joint distribution range of bone age and Risser sign, and an intervention measure set with intensity level is generated. The set of intervention measures with intensity levels is structured and arranged according to time dimension and execution sequence to generate structured individual early intervention plans.

[0013] Secondly, the present invention provides a system for generating early intervention plans for scoliosis based on multidisciplinary collaboration, including a standardization module for collecting multi-source structured patient feature data and performing standardization processing to generate a standardized patient feature dataset. The execution module performs multi-dimensional assessments on a standardized patient characteristic dataset and generates a multidisciplinary assessment result set. The mapping module, based on a multidisciplinary assessment result set, obtains the current stage of scoliosis development of the target patient, and assigns a decision-making role to each discipline at the current stage according to a predefined discipline role mapping relationship of development stage, forming a discipline role allocation structure. The aggregation module generates professional intervention suggestions for the corresponding disciplines based on the disciplinary role allocation structure and the multidisciplinary assessment results set, and aggregates them to form a multidisciplinary intervention suggestion set; The identification module, based on a multidisciplinary intervention suggestion set and a disciplinary role allocation structure, identifies the conflict types between different disciplinary intervention suggestions according to predefined disciplinary decision boundary rules, and performs collaborative iterative adjustments to conflicting suggestions according to role priority rules to generate a multidisciplinary collaborative consensus scheme. The intervention program module utilizes the individual environmental adaptation characteristics of target patients to screen the feasibility and intensity of multidisciplinary collaborative consensus programs, generating structured individual early intervention programs.

[0014] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, it implements any step of the method for generating an early intervention program for scoliosis based on multidisciplinary collaboration as described in the first aspect of the present invention.

[0015] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the method for generating an early intervention program for scoliosis based on multidisciplinary collaboration as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: By collecting and standardizing multi-source patient characteristic data, a multi-disciplinary assessment result set is generated through multi-dimensional specialist assessment. An improved spectral clustering algorithm is used to construct an asymmetric mapping map. Based on the growth potential partitioning logic, the development stages of scoliosis are dynamically divided and assigned disciplinary roles, forming a disciplinary role allocation structure. Combined with the manifold space projection and masked gradient descent adjustment of the interdisciplinary contraindication map, the hard boundary crossing and flexible boundary overlap conflicts between intervention suggestions from different disciplines are identified and resolved, generating a multi-disciplinary collaborative consensus scheme. Through the feasibility screening and intensity grading of individual environmental adaptation features, a structured individual early intervention scheme is produced. While avoiding the limitations of fixed weights and manual negotiation, it achieves automatic arbitration of topological constraints, dynamic role definition, and accurate generation of personalized executable schemes. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart illustrating a method for generating early intervention programs for scoliosis based on multidisciplinary collaboration.

[0019] Figure 2 This is a schematic diagram of a system for generating early intervention programs for scoliosis based on multidisciplinary collaboration. Detailed Implementation

[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0021] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0022] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0023] Reference Figures 1-2 This is one embodiment of the present invention, which provides a method for generating an early intervention plan for scoliosis based on multidisciplinary collaboration, including the following steps: S1. Collect multi-source structured patient feature data and perform standardization processing to generate a standardized patient feature dataset.

[0024] S1.1 Collect the target patient's age, gender, bone age, spinal imaging, body posture, motor behavior, and daily living behavior information to generate multi-source structured patient feature data. Perform missing value imputation and dimensional unification processing to generate multi-source structured patient feature data to be standardized.

[0025] Furthermore, missing values ​​are imputed, including using historical patient statistical characteristics or nearest neighbor interpolation, applying scaling uniformity to continuous variables, and uniform coding to categorical variables. This maps feature data from different sources and with different scales to a unified scale and format, thereby generating standardized multi-source structured patient feature data. This process achieves comparability between different types of features through data cleaning and scaling uniformity, enabling multidisciplinary assessments to be quantified and semantically interpreted based on unified data.

[0026] S1.2. Standardize multi-source structured patient feature data by performing format conversion and semantic encoding processing according to unified clinical data standards to generate a standardized patient feature dataset.

[0027] Furthermore, the standardized multi-source structured patient feature data is format-converted according to a unified clinical data standard, mapping various features to a standardized clinical data structure and performing unified semantic encoding processing. For example, spinal imaging parameters, postural quantitative indicators, and motor behavior quantitative data are mapped to clinical classification labels or quantitative levels, realizing the expression of multi-source information in a unified semantic space and generating a standardized patient feature dataset.

[0028] S2. Perform multi-dimensional assessments on standardized patient characteristic datasets to generate a multidisciplinary assessment result set.

[0029] S2.1 Based on the standardized patient characteristic dataset, the orthopedic assessment rules, rehabilitation medicine assessment rules, sports medicine assessment rules, growth and development assessment rules, and psychological and behavioral assessment rules are invoked respectively to generate preliminary specialty assessment results.

[0030] Furthermore, targeted analysis and processing of patient characteristics across different dimensions are conducted. Specifically, orthopedic assessment rules identify scoliosis morphological characteristics, progression risk characteristics, and structural stability characteristics based on spinal imaging information, bone age information, and postural information, forming orthopedic assessment results; rehabilitation medicine assessment rules identify muscle balance status, postural control status, and functional compensation status based on postural information and motor behavior information, forming rehabilitation medicine assessment results; sports medicine assessment rules analyze exercise load level, exercise risk level, and exercise adaptability based on motor behavior information and daily living behavior information, forming sports medicine assessment results; growth and development assessment rules analyze growth maturity status, growth potential status, and developmental stage status based on age information, gender information, and bone age information, forming growth and development assessment results; and psychological and behavioral assessment rules analyze compliance status, behavioral execution ability status, and psychological adaptation status based on daily living behavior information and behavioral characteristics information, forming psychological and behavioral assessment results.

[0031] S2.2. The preliminary specialty assessment results are encoded according to the unified semantic tags and classification coding rules to generate coded specialty assessment results, and then aggregated to generate a multi-disciplinary assessment result set.

[0032] Furthermore, the orthopedic assessment results, rehabilitation medicine assessment results, sports medicine assessment results, growth and development assessment results, and psychological and behavioral assessment results are converted into a unified expression format and assigned corresponding semantic labels and classification codes. Specifically, structural abnormality assessment results are assigned a structural risk label, functional abnormality assessment results are assigned a functional risk label, growth and development assessment results are assigned a growth status label, behavioral and psychological assessment results are assigned a behavioral status label, and motor adaptation assessment results are assigned a motor status label. After coding, coded specialty assessment results are formed. These coded specialty assessment results are then aggregated according to the patient's unique identifier. The assessment labels and classification codes from different disciplines are linked, and the results are summarized according to a unified data structure to form a multidisciplinary assessment result set.

[0033] S3. Based on the multidisciplinary assessment results set, obtain the current stage of scoliosis development of the target patient, and assign a decision-making role to each discipline at the current stage according to the predefined discipline role mapping relationship of development stage, forming a discipline role allocation structure.

[0034] S3.1 Based on the multidisciplinary assessment results set, extract the Cobb angle difference, bone age grading label and secondary sexual characteristic development label at continuous time points, input them into the growth potential partitioning logic, and divide the growth active area, slowing area and stable area by comparing the joint distribution interval of bone age and Risser sign to obtain the current stage of scoliosis development of the target patient.

[0035] Furthermore, bone age grading labels and secondary sexual characteristic development labels are directly read from the standardized patient feature dataset. Cobb angle differences are used to characterize the progression trend of scoliosis, bone age grading labels to characterize skeletal maturity, and secondary sexual characteristic development labels to characterize the pubertal development process. The Cobb angle differences, bone age grading labels, and secondary sexual characteristic development labels at continuous time points are input into the growth potential partitioning logic. The growth potential partitioning logic compares the combined distribution intervals of bone age and Risser sign. Areas with strong growth potential and high sensitivity to scoliosis progression are classified as active growth zones; areas with declining growth rate and slowing scoliosis progression are classified as slowing growth zones; and areas with high skeletal maturity and reduced risk of scoliosis progression are classified as stable zones. The dynamic progression is reflected by the Cobb angle differences at continuous time points, the developmental background is reflected by the bone age grading labels and secondary sexual characteristic development labels, and the stage determination is completed by the combined distribution intervals of bone age and Risser sign, thus obtaining the current scoliosis development stage of the target patient.

[0036] The expression for the angle difference is: ; in, For the target patient in the first The Cobb angle of the principal curve was measured during the follow-up imaging evaluation. The value of the Cobb angle of the main curve is immediately adjacent to the value measured in the previous follow-up visit. Indicates the first The Cobb angle difference at consecutive time points reflects the unidirectional change in the scoliosis angle between two adjacent assessments.

[0037] Specifically, the Cobb angle difference at consecutive time points, bone age grading labels, and secondary sexual characteristic development labels jointly participate in the growth potential zoning logic, which is jointly defined by structural progress, skeletal maturation, and developmental processes. For example, when the Cobb angle changes rapidly but bone age tends to mature, the growth potential zoning logic can avoid mechanically classifying the progression trend into a high-risk growth period; when the Cobb angle changes relatively slowly but bone age and Risser sign indicate that it is still in a rapid growth period, the growth potential zoning logic can retain potential progression risks. The combined distribution range of bone age and Risser sign can express the correspondence between skeletal maturation status and the sensitive period of scoliosis progression, making the growth active zone, slowing zone, and stable zone clinically interpretable.

[0038] S3.2. Based on the current stage of scoliosis development of the target patient, we traverse the time sequence records of multidisciplinary joint interventions under different progression trajectories in the historical scoliosis case database, and construct an undirected weighted network with disciplines as nodes and the correlation of intervention effectiveness as edges.

[0039] Furthermore, based on how to traverse the time-series records of multidisciplinary joint interventions under different progression trajectories in the historical scoliosis case database, this study extracts the participation order, intervention subjects, intervention results, and collaborative relationships of each discipline at different intervention stages. Orthopedics, rehabilitation medicine, sports medicine, growth and development, and psychological behavior are each designated as a discipline node. The correlation between the joint intervention and the intervention effectiveness is used as edges, and edge weights are assigned based on the correlation of intervention effectiveness. The correlation of intervention effectiveness originates from the collaborative contribution relationship of different disciplines in the same progression trajectory, reflecting the strength of cooperation between disciplines at a specific stage of scoliosis development. An undirected weighted network is constructed with disciplines as nodes and the correlation of intervention effectiveness as edges.

[0040] Specifically, the time-series records of multidisciplinary collaborative interventions under different progression trajectories in the historical scoliosis case database can preserve the temporal order and effectiveness correlation of interdisciplinary collaboration, ensuring that interdisciplinary relationships are not static empirical arrangements but rather derived from the collaborative results in the actual progression trajectory. For example, in the active growth zone, orthopedic interventions and rehabilitation medicine interventions may show a stronger effectiveness correlation, while in the stable zone, rehabilitation medicine interventions and sports medicine interventions may show a stronger effectiveness correlation. An undirected weighted network with disciplines as nodes and intervention effectiveness correlation as edges can transform multidisciplinary collaborative intervention relationships into a segmentable and comparable network structure, enabling subsequent improved spectral clustering algorithms to identify interdisciplinary collaborative clustering relationships at different stages based on network connection strength.

[0041] S3.3. The network is segmented using an improved spectral clustering algorithm. The improved spectral clustering algorithm introduces stage labels as constraints after the Laplace matrix eigenvalue decomposition to force the differentiation of subject cluster centers at different growth stages, generate subject role prototypes at each stage, and form an asymmetric mapping spectrum.

[0042] Furthermore, an improved spectral clustering algorithm is used to segment the undirected weighted network with disciplines as nodes and the correlation between intervention effectiveness and other factors as edges. The improved spectral clustering algorithm first forms a Laplace matrix based on the undirected weighted network and then performs eigenvalue decomposition on the Laplace matrix to obtain feature representations that reflect the strength of the correlation between effectiveness among discipline nodes. Subsequently, stage labels are added as constraints to the clustering process after eigenvalue decomposition, forcibly distinguishing the discipline cluster centers corresponding to active growth areas, mitigation areas, and stable areas, avoiding mixed classification of discipline collaboration relationships under different growth stages. After network segmentation, discipline role prototypes for each stage are formed. These prototypes express the tendency of each discipline to assume a leading, supporting, constraining, or auxiliary role in different stages of scoliosis development. The discipline role prototypes for each stage are arranged according to the development stage and role relationship to form an asymmetric mapping graph.

[0043] Specifically, the improved spectral clustering algorithm utilizes stage labels as constraints after Laplacian matrix eigenvalue decomposition, ensuring that the subject clustering results are simultaneously constrained by network effectiveness correlation and the scoliosis development stage. Ordinary spectral clustering tends to group subject nodes into similar groups based on overall correlation strength, while stage labels as constraints can forcibly distinguish subject clustering centers at different growth stages. For example, active growth areas may highlight the dominant tendency of orthopedics in structural risk control, while stable areas may highlight the dominant tendency of rehabilitation medicine in functional maintenance. The value of asymmetric mapping lies in the fact that the relationship between development stage and subject role is not a two-way equivalence; the same subject can assume different decision-making roles at different stages, and the same role can be assumed by different subjects at different stages, thereby improving the adaptability of subject role allocation to stage differences.

[0044] S3.4 Input the high-risk compensation mode labels from the multidisciplinary assessment result set into the path matching interface of the asymmetric mapping graph. Perform conflict resolution and boundary calibration on the initial role combination through topological sorting within the graph to obtain the decision-making role of each discipline at the current stage.

[0045] Furthermore, the asymmetric mapping graph searches for corresponding paths in the subject role prototypes at each stage based on high-risk compensation mode labels, obtaining initial role combinations. These initial role combinations are then sorted through internal topological sorting of the graph, establishing clear sequence and boundary relationships among leading, supporting, constraining, and auxiliary roles. To address potential role overlaps, overlapping authority, or unclear intervention boundaries in the initial role combinations, conflict resolution and boundary calibration are performed based on the topological sorting results, ensuring consistency between the responsibilities and decision-making priorities of each subject at the current stage. After completing path matching, topological sorting, conflict resolution, and boundary calibration, the decision-making role of each subject at the current stage is obtained.

[0046] Specifically, the high-risk compensation pattern label serves as the path matching basis for the asymmetric mapping map, ensuring that the allocation of disciplinary roles is not only influenced by the current stage of scoliosis development in the target patient but also reflects individualized compensation risks. For example, when the risk of structural progression is prominent, the path matching results can strengthen the priority of orthopedic decisions; when the risk of functional compensation is prominent, the path matching results can strengthen the boundaries of rehabilitation medicine interventions; and when the risk of exercise load is prominent, the path matching results can increase the weight of the sports medicine constraint role. The topological ordering within the map can transform role relationships into ordered decision chains, allowing role conflicts in the initial role combination to be resolved through priority relationships. Conflict resolution and boundary calibration can prevent multiple disciplines from repeatedly dominating or contradicting each other on the same intervention.

[0047] S3.5 Summarize the decision-making roles of disciplines at the current stage and establish an index association with the current stage of scoliosis development of the target patient to form a discipline role allocation structure.

[0048] Furthermore, the decision-making roles of each discipline at the current stage are summarized, with the decision-making roles corresponding to orthopedics, rehabilitation medicine, sports medicine, growth and development, and psychological behavior recorded separately, and an index is established to associate them with the current stage of scoliosis development in the target patient. This index is used to express the correspondence between the current stage of scoliosis development in the target patient and the decision-making roles of each discipline. After completing the summary of discipline decision-making roles and the index association of development stages, a discipline role allocation structure is formed.

[0049] S4. Based on the subject role allocation structure and the multidisciplinary assessment results set, generate professional intervention suggestions for the corresponding subjects, and summarize them to form a multidisciplinary intervention suggestion set.

[0050] S4.1 Based on the decision-making roles and authority boundaries defined in the subject role allocation structure, and combined with the quantitative indicators in the multidisciplinary evaluation result set, generate professional intervention suggestions for the corresponding subjects.

[0051] Furthermore, for orthopedics, based on structural risk indicators, Cobb angle trends, and growth potential status, professional recommendations are generated, including orthopedic brace wearing plans, necessary imaging follow-up frequencies, and surgical intervention assessments. For rehabilitation medicine, based on muscle balance status, postural control ability, and compensation mode indicators, functional training plans, postural correction plans, and rehabilitation intervention cycle recommendations are generated. For sports medicine, based on exercise load indicators, exercise risk and adaptability indicators, exercise type selection, intensity and frequency planning, and exercise risk control measures are generated. For growth and development, based on bone age grading and developmental potential indicators, growth monitoring plans and intervention priority adjustment recommendations are generated. For psychobehavioral medicine, based on compliance status and behavioral execution ability indicators, behavioral guidance, compliance improvement, and psychological intervention recommendations are generated. Each discipline's professional intervention recommendations strictly adhere to its authority boundaries, ensuring that the content of the recommendations is consistent with its role and responsibilities, forming a set of professional intervention recommendations for the corresponding discipline.

[0052] S4.2. The professional intervention suggestions of the corresponding disciplines are structured and encapsulated according to the unified clinical semantic coding rules to generate encapsulated discipline intervention suggestions, and then aggregated and indexed to generate a multidisciplinary intervention suggestion set.

[0053] Furthermore, each intervention is mapped to an indexable set of codes, tags, and parameters, including intervention type, execution goal, priority, and intensity level. After encapsulation, intervention recommendations from different disciplines are aggregated to create a set of intervention recommendations indexed by patient identification and intervention stage. Simultaneously, multidimensional indexes are constructed for the intervention recommendations, including discipline indexes, intervention type indexes, and execution priority indexes, generating a multidisciplinary intervention recommendation set.

[0054] S5. Based on the multidisciplinary intervention suggestion set and the disciplinary role allocation structure, and according to the predefined disciplinary decision boundary rules, identify the conflict types between different disciplinary intervention suggestions, and adjust the conflict suggestions in a collaborative iterative manner according to the role priority rules to generate a multidisciplinary collaborative consensus scheme.

[0055] S5.1 Based on the multidisciplinary intervention suggestion set and the subject role allocation structure, a predefined subject decision boundary rule is generated by constructing an interdisciplinary taboo graph. The interdisciplinary taboo graph uses a hypergraph structure to characterize the asymmetric exclusion relationship between the absolute taboo domain of orthopedics and the relatively safe domain of rehabilitation medicine. Based on the topological distance of the exclusion relationship, hard boundaries and flexible boundaries are divided to establish the overlapping operational range of different subject suggestions.

[0056] Furthermore, the interdisciplinary taboo map utilizes a hypergraph structure to represent the asymmetric exclusion relationships between interventions from orthopedics, rehabilitation medicine, sports medicine, growth and development, and psychological and behavioral disciplines. The absolute taboo domain in orthopedics defines high-risk operation areas, while the relatively safe domain in rehabilitation medicine defines adjustable operation areas. Hard and soft boundaries are delineated using the topological distance of the exclusion relationships, clarifying the overlapping operation areas of intervention recommendations from different disciplines. Hard boundaries are used to identify areas where interventions absolutely cannot overlap, while soft boundaries are used to identify areas where fine-tuning and partial overlap are permissible. This creates clear operational constraints among multidisciplinary intervention recommendations, enabling safe collaboration within these boundaries.

[0057] Specifically, by constructing an interdisciplinary taboo map, the complex asymmetric exclusion relationships between disciplinary intervention recommendations are visualized and structured, enabling intervention conflicts to be precisely quantified and determined within a topological space. The division between hard and flexible boundaries not only reflects the differences between absolute prohibitions and reconcilable differences between disciplines but also provides overlapping operational zones for interventions, allowing different disciplines to maintain coordination within a shared intervention space. For example, the restrictions on spinal structure in orthopedics and the permissible fine-tuning of functional training in rehabilitation medicine can be differentiated and integrated through topological distance. Representing asymmetric exclusion relationships through a hypergraph structure gives multidisciplinary intervention boundary rules the characteristics of quantification, operability, and scalability, providing a structured basis for conflict detection and collaborative iterative adjustments.

[0058] S5.2 Input the action amplitude, execution frequency and contraindication range from the multidisciplinary intervention suggestion set into the conflict detection engine based on the interdisciplinary contraindication map. By calculating the projection position of the action amplitude, execution frequency and contraindication range parameter vectors in the map manifold space, identify the conflict type currently existing in the target patient. The conflict type includes hard boundary crossing conflict and flexible boundary overlapping conflict.

[0059] Furthermore, the parameter vectors of action amplitude, execution frequency, and contraindication range from the multidisciplinary intervention suggestion set are input into a conflict detection engine based on a cross-disciplinary contraindication map. By projecting the parameter vector of each intervention action onto the map manifold space, the relationship between action amplitude, execution frequency, and contraindication range and the hard and soft boundaries of the map is calculated to identify the current conflict type of the target patient, including hard boundary overstepping conflict and soft boundary overlapping conflict. Hard boundary overstepping conflict indicates that the intervention operation parameters exceed the disciplinary decision boundary, while soft boundary overlapping conflict indicates that the intervention operation overlaps within the allowable range but requires coordination and fine-tuning. Projection calculation can quantify the degree of deviation of the intervention action from the boundary, achieving accurate identification of potential conflicts between multidisciplinary intervention suggestions.

[0060] Specifically, by mapping the amplitude, frequency, and contraindications of intervention actions to the manifold space of an interdisciplinary contraindication map, multidisciplinary intervention conflicts are transformed from abstract semantic relationships into quantifiable spatial projection problems, enabling precise identification of intervention conflict types. The determination of hard boundary crossing conflicts and flexible boundary overlapping conflicts allows intervention actions to be classified according to the strictness of the boundaries. For example, orthopedic recommendations for brace wearing angles exceeding hard boundaries can be directly identified as boundary crossing conflicts, while rehabilitation medicine training actions overlapping within flexible boundaries can be identified as adjustable conflicts. This transforms multidisciplinary intervention conflict detection from experience-based judgment to quantitative analysis, supporting safe and controllable collaborative iterative adjustments and improving the feasibility and safety of multidisciplinary collaborative interventions.

[0061] Intervention recommendations from different disciplines are transformed into feature vectors with a unified dimension, and a specific intervention action, such as the recommended parameter vector for spinal rotation breathing training, is defined.

[0062] The suggested parameter vector expression is: ; in, For the proposed parameter vector, For the amplitude of the action, For execution frequency, This is within the prohibited scope.

[0063] The topological structure of the interdisciplinary taboo graph is represented as follows: ; in, For the topological structure of interdisciplinary taboo graphs As a disciplinary node, As a hard boundary, It is a flexible boundary.

[0064] The collision detection engine calculates the proposed parameter vector on the Riemannian manifold. Projection position on .

[0065] ; in, For the projection position, For projection operators, For interdisciplinary taboo graphs Riemannian manifold space Any point in the array.

[0066] S5.3 Input the conflict type and subject role allocation structure into the collaborative iterative adjustment logic based on role priority. The collaborative iterative adjustment logic adopts the coordinate gradient descent method with mask. During the iteration process, the suggestion parameter coordinates corresponding to the dominant role are frozen. Fine-tuning is applied to the overlapping parameters of the supporting role and the constraint role along the negative gradient direction of the graph, and the parameters are returned to the overlapping operation range to generate the adjusted multidisciplinary intervention suggestion set.

[0067] Furthermore, a masked coordinate gradient descent method is employed for the adjustment process. The coordinates of the intervention suggestion parameters corresponding to the dominant role are frozen during iteration, and only the overlapping parameters of the supporting and constraining roles are fine-tuned along the negative gradient direction of the graph. During fine-tuning, the amplitude, execution frequency, and taboo range parameters of each intervention action are locally adjusted according to the conflict type, bringing the adjusted parameters back to the predefined overlapping operation range. This preserves the priority and integrity of the dominant role's intervention effect, gradually optimizing the intervention parameters of the supporting and constraining roles to achieve consistency with the dominant role's intervention while avoiding hard boundary conflicts. This generates an adjusted multidisciplinary intervention suggestion set, ensuring safe collaboration among the various disciplines under role priority constraints.

[0068] Specifically, a masked coordinate gradient descent method is used to differentiate the intervention parameters of the dominant role from those of other roles. This prioritizes the decision-making weight of the dominant role, while the parameters of supporting and constraining roles are fine-tuned within the adjustable space using negative gradient directions to resolve conflicts. For example, the brace wearing angle, which is dominated by orthopedics, is frozen, while the posture training movements, which are dominated by rehabilitation medicine, are fine-tuned within flexible boundaries to ensure compatibility between functional training and structural protection. Through the mapping relationship between role priority and parameter space, multidisciplinary conflicts are transformed from a semantic level into a continuously optimizable parameter adjustment problem, enabling collaborative optimization of multidisciplinary intervention recommendations under the conditions of ensuring safety boundaries and prioritizing the dominant role.

[0069] S5.4. Input the adjusted set of multidisciplinary intervention suggestions back into the interdisciplinary taboo graph for boundary re-verification, and perform conflict detection and collaborative iterative adjustment repeatedly until there are no hard boundary crossing conflicts, and generate a multidisciplinary collaborative consensus scheme.

[0070] Furthermore, a closed-loop optimization mechanism is formed through cyclic conflict detection and collaborative iterative adjustment, enabling multidisciplinary intervention recommendations to continuously approach the optimal collaborative state under both hard and flexible boundary constraints. For example, when rehabilitation medicine intervention and sports medicine intervention still have slight overlap within the flexible boundary, iterative adjustments are made to converge their parameters to a safe overlap range without affecting the orthopedic-led intervention. This upgrades static conflict identification to a dynamic iterative optimization process, achieving safe integration of multidisciplinary collaborative interventions under complex boundary conditions and ensuring the feasibility of the final multidisciplinary collaborative consensus scheme. S6. Utilize the individual environmental adaptation characteristics of target patients to conduct feasibility screening and intensity grading of multidisciplinary collaborative consensus programs, and generate structured individual early intervention programs.

[0071] S6.1. Utilize the individual environmental adaptation characteristics of the target patient to perform family executive capacity matching tests on the intervention measures in the multidisciplinary collaborative consensus plan, and generate a set of intervention measures after feasibility filtering.

[0072] Furthermore, by assessing available family resources, patient compliance, and environmental constraints, interventions are matched with the family's ability to implement them. For each intervention, its implementation complexity, required assistive devices, monitoring needs, and compliance requirements are analyzed, and interventions that can be safely and effectively implemented in the target patient's home environment are selected, forming a set of interventions after feasibility filtering.

[0073] S6.2 Input the set of intervention measures after feasibility filtering and the growth potential information of the target patient into the intensity grading logic, and discretize the intervention intensity according to the joint distribution range of bone age and Risser sign to generate a set of intervention measures with intensity level.

[0074] Furthermore, by referencing the combined distribution range of bone age and Risser sign, the intensity of each intervention measure is discretized and calibrated, including the grading of intervention amplitude, frequency, and training intensity. Based on growth potential status, intervention measures are divided into different intensity levels, allowing for relatively high-intensity interventions during high-potential periods and low-intensity or maintenance interventions during low-potential periods, thus generating a set of intervention measures with intensity levels.

[0075] S6.3 Arrange the set of intervention measures with intensity levels in a structured manner according to the time dimension and the execution order to generate a structured individual early intervention plan.

[0076] Furthermore, the set of intervention measures with intensity levels is structured and arranged according to time and execution sequence, including intervention start and end times, execution cycle, daily execution sequence, and coordination relationships between various disciplines. By comprehensively considering intervention priorities, intensity levels, and family feasibility, the intervention measures are formed into a complete sequence in the time dimension, ensuring that different disciplines do not conflict within the same stage, while maintaining the intervention rhythm consistent with the developmental stage, thus generating a structured individualized early intervention plan.

[0077] This embodiment also provides a system for generating early intervention plans for scoliosis based on multidisciplinary collaboration, including: The standardization module collects multi-source structured patient feature data and performs standardization processing to generate a standardized patient feature dataset; The execution module performs multi-dimensional assessments on a standardized patient characteristic dataset and generates a multidisciplinary assessment result set. The mapping module, based on a multidisciplinary assessment result set, obtains the current stage of scoliosis development of the target patient, and assigns a decision-making role to each discipline at the current stage according to a predefined discipline role mapping relationship of development stage, forming a discipline role allocation structure. The aggregation module generates professional intervention suggestions for the corresponding disciplines based on the disciplinary role allocation structure and the multidisciplinary assessment results set, and aggregates them to form a multidisciplinary intervention suggestion set; The identification module, based on a multidisciplinary intervention suggestion set and a disciplinary role allocation structure, identifies the conflict types between different disciplinary intervention suggestions according to predefined disciplinary decision boundary rules, and performs collaborative iterative adjustments to conflicting suggestions according to role priority rules to generate a multidisciplinary collaborative consensus scheme. The intervention program module utilizes the individual environmental adaptation characteristics of target patients to screen the feasibility and intensity of multidisciplinary collaborative consensus programs, generating structured individual early intervention programs.

[0078] This embodiment also provides a computer device applicable to the method for generating early intervention plans for scoliosis based on multidisciplinary collaboration, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the method for generating early intervention plans for scoliosis based on multidisciplinary collaboration as proposed in the above embodiment.

[0079] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0080] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the method for generating an early intervention program for scoliosis based on multidisciplinary collaboration as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0081] In summary, this invention collects and standardizes multi-source patient characteristic data, generates a multidisciplinary assessment result set through multi-dimensional specialist evaluation, constructs an asymmetric mapping map using an improved spectral clustering algorithm, dynamically divides the development stages of scoliosis according to growth potential partitioning logic and assigns disciplinary roles, forming a disciplinary role allocation structure. Combined with manifold space projection and masked gradient descent adjustment of the interdisciplinary contraindication map, it identifies and resolves conflicts between hard boundary crossings and flexible boundary overlaps between intervention suggestions from different disciplines, generating a multidisciplinary collaborative consensus scheme. Through feasibility screening and intensity grading of individual environmental adaptation features, it produces structured individual early intervention schemes. While avoiding the limitations of fixed weights and manual negotiation, it achieves automatic arbitration of topological constraints, dynamic role definition, and accurate generation of personalized executable schemes.

[0082] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for generating an early intervention program for scoliosis based on multi-disciplinary collaboration, characterized in that: include, Collect multi-source structured patient feature data and perform standardization processing to generate a standardized patient feature dataset; Perform multidimensional assessments on standardized patient characteristic datasets to generate a multidisciplinary assessment result set; Based on the multidisciplinary assessment results set, the current stage of scoliosis development of the target patient is obtained. According to the predefined development stage discipline role mapping relationship, each discipline is assigned a decision-making role under the current stage, forming a discipline role allocation structure. Based on the disciplinary role allocation structure and the multidisciplinary assessment results set, professional intervention suggestions for the corresponding disciplines are generated and summarized to form a multidisciplinary intervention suggestion set; Based on a multidisciplinary set of intervention suggestions and a disciplinary role allocation structure, and in accordance with predefined disciplinary decision boundary rules, conflict types among different disciplinary intervention suggestions are identified. According to role priority rules, conflicting suggestions are collaboratively and iteratively adjusted to generate a multidisciplinary collaborative consensus scheme. By leveraging the individual environmental adaptation characteristics of target patients, we can screen the feasibility and intensity of multidisciplinary collaborative consensus programs and generate structured individualized early intervention programs.

2. The multi-disciplinary collaboration based scoliosis early intervention program generation method of claim 1, wherein: The standardized patient feature dataset includes, Collect age, gender, bone age, spinal imaging, body shape, motor behavior, and daily living behavior information of the target patients to generate multi-source structured patient feature data. Perform missing value imputation and dimensional unification processing to generate multi-source structured patient feature data to be standardized. Standardized multi-source structured patient feature data is processed by format conversion and semantic encoding according to unified clinical data standards to generate a standardized patient feature dataset.

3. The method for generating an early intervention plan for scoliosis based on multidisciplinary collaboration as described in claim 2, characterized in that: The multidisciplinary assessment results set includes, Based on a standardized patient characteristic dataset, preliminary specialist assessment results are generated by calling orthopedic assessment rules, rehabilitation medicine assessment rules, sports medicine assessment rules, growth and development assessment rules, and psychological and behavioral assessment rules respectively. The preliminary specialty assessment results are encoded according to unified semantic tags and classification coding rules to generate coded specialty assessment results, and then aggregated to generate a multidisciplinary assessment result set.

4. The method for generating an early intervention plan for scoliosis based on multidisciplinary collaboration as described in claim 3, characterized in that: The subject-specific role allocation structure includes: Based on the multidisciplinary assessment results set, the Cobb angle difference, bone age grading label and secondary sexual characteristic development label at continuous time points are extracted and input into the growth potential partitioning logic. By comparing the joint distribution interval of bone age and Risser sign, the growth active area, slowing area and stable area are divided to obtain the current stage of scoliosis development of the target patient. To determine the current stage of scoliosis development in the target patient, we traversed the time-series records of multidisciplinary joint interventions under different progression trajectories in the historical scoliosis case database and constructed an undirected weighted network with disciplines as nodes and the correlation of intervention effectiveness as edges. An improved spectral clustering algorithm is used to segment the network. The improved spectral clustering algorithm introduces stage labels as constraints after the Laplacian matrix eigenvalue decomposition to force the differentiation of subject clustering centers at different growth stages, generate subject role prototypes at each stage, and form an asymmetric mapping spectrum. The high-risk compensation pattern labels in the multidisciplinary assessment result set are input into the path matching interface of the asymmetric mapping graph. The initial role combination is resolving conflicts and calibrating boundaries through topological sorting within the graph to obtain the decision-making role of each discipline at the current stage. The decision-making roles of various disciplines at the current stage are summarized and indexed to establish an association between the target patient's current stage of scoliosis development, thus forming a structure for assigning disciplinary roles.

5. The method for generating an early intervention plan for scoliosis based on multidisciplinary collaboration as described in claim 4, characterized in that: The set of multidisciplinary intervention recommendations includes, Based on the decision-making roles and authority boundaries defined in the subject-specific role allocation structure, and combined with the quantitative indicators in the multidisciplinary evaluation result set, professional intervention suggestions for the corresponding subjects are generated. The professional intervention suggestions for the corresponding disciplines are structured and encapsulated according to unified clinical semantic coding rules to generate encapsulated discipline intervention suggestions. These suggestions are then aggregated and indexed to generate a multidisciplinary intervention suggestion set.

6. The method for generating an early intervention plan for scoliosis based on multidisciplinary collaboration as described in claim 5, characterized in that: The multidisciplinary collaborative consensus scheme includes, Based on a multidisciplinary set of intervention recommendations and a disciplinary role allocation structure, a predefined disciplinary decision boundary rule is generated by constructing an interdisciplinary taboo graph. The interdisciplinary taboo graph uses a hypergraph structure to characterize the asymmetric exclusion relationship between the absolute taboo domain of orthopedics and the relatively safe domain of rehabilitation medicine. Hard and soft boundaries are divided according to the topological distance of the exclusion relationship, and the overlapping operational range of different disciplinary recommendations is established. The action amplitude, execution frequency, and contraindication range from the multidisciplinary intervention suggestion set are input into a conflict detection engine based on an interdisciplinary contraindication map. By calculating the projection position of the parameter vectors of action amplitude, execution frequency, and contraindication range in the map manifold space, the conflict type currently existing in the target patient is identified. The conflict type includes hard boundary crossing conflict and flexible boundary overlapping conflict. The conflict type and subject role allocation structure are input into the collaborative iterative adjustment logic based on role priority. The collaborative iterative adjustment logic adopts the coordinate gradient descent method with mask. During the iteration process, the suggestion parameter coordinates corresponding to the dominant role are frozen. The overlapping parameters of the supporting role and the constraint role are fine-tuned along the negative gradient direction of the graph and returned to the overlapping operation range to generate the adjusted multidisciplinary intervention suggestion set. The adjusted set of multidisciplinary intervention suggestions is then input into the interdisciplinary taboo graph for boundary re-verification. Conflict detection and collaborative iterative adjustment are performed repeatedly until there are no hard boundary crossing conflicts, generating a multidisciplinary collaborative consensus scheme.

7. The method for generating an early intervention plan for scoliosis based on multidisciplinary collaboration as described in claim 6, characterized in that: The structured individual early intervention program includes, By utilizing the individual environmental adaptation characteristics of target patients, the family executive capacity matching degree of intervention measures in the multidisciplinary collaborative consensus plan is tested, and a set of intervention measures after feasibility filtering is generated. The set of intervention measures after feasibility filtering and the growth potential information of the target patients are input into the intensity grading logic. The intervention intensity is discretized and calibrated according to the joint distribution range of bone age and Risser sign, and an intervention measure set with intensity level is generated. The set of intervention measures with intensity levels is structured and arranged according to time dimension and execution sequence to generate structured individual early intervention plans.

8. A system for generating early intervention plans for scoliosis based on multidisciplinary collaboration, based on the method for generating early intervention plans for scoliosis based on multidisciplinary collaboration as described in any one of claims 1 to 7, characterized in that: This includes a standardization module, which collects multi-source structured patient feature data and performs standardization processing to generate a standardized patient feature dataset; The execution module performs multi-dimensional assessments on a standardized patient characteristic dataset and generates a multidisciplinary assessment result set. The mapping module, based on a multidisciplinary assessment result set, obtains the current stage of scoliosis development of the target patient, and assigns a decision-making role to each discipline at the current stage according to a predefined discipline role mapping relationship of development stage, forming a discipline role allocation structure. The aggregation module generates professional intervention suggestions for the corresponding disciplines based on the disciplinary role allocation structure and the multidisciplinary assessment results set, and aggregates them to form a multidisciplinary intervention suggestion set; The identification module, based on a multidisciplinary intervention suggestion set and a disciplinary role allocation structure, identifies the conflict types between different disciplinary intervention suggestions according to predefined disciplinary decision boundary rules, and performs collaborative iterative adjustments to conflicting suggestions according to role priority rules to generate a multidisciplinary collaborative consensus scheme. The intervention program module utilizes the individual environmental adaptation characteristics of target patients to screen the feasibility and intensity of multidisciplinary collaborative consensus programs, generating structured individual early intervention programs.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the method for generating an early intervention program for scoliosis based on multidisciplinary collaboration as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the method for generating an early intervention program for scoliosis based on multidisciplinary collaboration as described in any one of claims 1 to 7.

Images