Intelligent Analysis System and Method for Rehabilitation Treatment of Lumbar Disc Herniation

Abstract

The present application discloses an intelligent analysis system and method for the rehabilitation treatment of lumbar disc herniation, which relates to the field of intelligent analysis. By using the semantic coding technology based on deep learning, the abstract clinical information such as the characteristics of the lesion site, the description of pain perception, and the motor function requirements is transformed into high-dimensional vectors, enabling the system to more accurately understand and simulate the patient's situation in a multi-dimensional space. Further, a dynamic mapping mechanism between the rehabilitation starting point and the goal is established to capture the non-linear correlation between the patient's current state and the ideal rehabilitation goal, and based on this, the intelligent derivation of the rehabilitation path is realized. Through the dynamic analysis of the patient's multi-dimensional information, the system can better understand the unique needs of each patient, improve the matching accuracy of the rehabilitation treatment plan, and thus make the treatment more personalized and effective.

Description

Technical Field

[0001] The present application relates to the field of intelligent analysis, and more specifically, to an intelligent analysis system and method for the rehabilitation treatment of lumbar disc herniation. Background Art

[0002] In the field of rehabilitation treatment of lumbar disc herniation, the formulation of traditional rehabilitation programs mainly relies on doctors' clinical experience and static evaluation indicators, lacking the ability to dynamically analyze multi-dimensional information of patients and adapt to individual needs. Existing technologies usually conduct single-dimensional evaluations through standardized scales or imaging features, making it difficult to effectively integrate unstructured text descriptions (such as lesion site characteristics, individualized expressions of pain perception) and structured data (such as age, medical history, etc.) in patients' medical records, resulting in insufficient matching accuracy between rehabilitation goals and treatment means.

[0003] In addition, due to the large differences in the specific conditions of each patient, such as lesion site, pain level, requirements for motor function recovery, and personal basic information, it is difficult to meet individual needs with a unified treatment plan, resulting in uneven treatment effects.

[0004] Therefore, an optimized intelligent analysis method for the rehabilitation treatment of lumbar disc herniation is expected. Summary of the Invention

[0005] To solve the above technical problems, the present application is proposed. The present application provides an intelligent analysis system and method for the rehabilitation treatment of lumbar disc herniation. By using semantic encoding technology based on deep learning, abstract clinical information such as lesion site characteristics, pain perception descriptions, and motor function requirements is transformed into high-dimensional vectors, enabling the system to more accurately understand and simulate the situation of patients in a multi-dimensional space. Further, a dynamic mapping mechanism between the rehabilitation starting point and the goal is established to capture the non-linear relationship between the patient's current state and the ideal rehabilitation goal, and based on this, the intelligent derivation of the rehabilitation path is realized. Through the dynamic analysis of patients' multi-dimensional information, the system can better understand the unique needs of each patient, improve the matching accuracy of rehabilitation treatment plans, and thus make the treatment more personalized and effective.

[0006] According to one aspect of the present application, an intelligent analysis method for the rehabilitation treatment of lumbar disc herniation is provided, which includes:

[0007] Obtaining the medical record of the target patient object, where the medical record includes a current condition part and a target part for motor function recovery requirements;

[0008] Extracting the semantic information of the medical record to obtain a set of semantic encoding vectors for motor function recovery requirements and the current condition part of the medical record;

[0009] The semantic coding vector of the motor function recovery demand is used as the rehabilitation target vector and the set of semantic coding vectors of the current condition is used as the set of rehabilitation starting point vectors. These vectors are input into the rehabilitation starting point-target semantic analysis network to obtain the rehabilitation starting point-target semantic dynamic response coding vector, including: extracting the deep implicit coding features of the rehabilitation target vector and the set of rehabilitation starting point vectors respectively, and constructing a local feature response matrix between the two based on a dynamic semantic matching mechanism to obtain a set of rehabilitation starting point-target local semantic feature response anchor coding matrices; and realizing multi-matrix weighted fusion through an adaptive weight allocation mechanism to generate the rehabilitation starting point-target semantic dynamic response coding vector.

[0010] Generate personalized rehabilitation plans based on the rehabilitation starting point-goal semantic dynamic response encoding vector.

[0011] Furthermore, the current condition section includes the patient's basic information, descriptions of the lesion site, and descriptions of the pain level; the motor function recovery demand target section includes descriptions of the motor function recovery demand.

[0012] Furthermore, the semantic information of the medical condition document is extracted to obtain a set of semantic coding vectors of motor function recovery requirements and semantic coding vectors of the current condition, including:

[0013] Perform semantic embedding coding on the description of the lesion site, the description of the pain level, the description of the motor function recovery requirements, and the patient's basic information to obtain a semantic coding vector for the lesion site description, a semantic coding vector for the pain level description, a semantic coding vector for the motor function recovery requirements, and a semantic coding vector for the patient's basic information;

[0014] The semantic coding vector describing the lesion site, the semantic coding vector describing the pain level and the semantic coding vector describing the patient's basic information are merged to obtain a set of semantic coding vectors describing the current condition of the disease.

[0015] Furthermore, the patient's basic information includes age, gender, weight and medical history.

[0016] Furthermore, the deep implicit coding features of the sets of rehabilitation target vectors and rehabilitation starting point vectors are extracted respectively, and a local feature response matrix between the two is constructed based on a dynamic semantic matching mechanism to obtain a set of rehabilitation starting point-target local semantic feature response anchor coding matrices, including:

[0017] Performing deep latent feature extraction based on fully connected coding on the rehabilitation target vector to obtain a deep latent coding vector of the rehabilitation target feature;

[0018] Performing deep latent feature extraction based on fully connected coding on each rehabilitation starting point vector in the set of rehabilitation starting point vectors to obtain a set of rehabilitation starting point feature deep latent coding vectors;

[0019] Perform rehabilitation start - goal semantic dynamic response encoding on each rehabilitation start - point feature deep implicit encoding vector in the set of rehabilitation goal feature deep implicit encoding vectors and rehabilitation start - point feature deep implicit encoding vectors to obtain a set of rehabilitation start - goal local semantic feature response anchor encoding matrices.

[0020] Furthermore, implement multi - matrix weighted fusion through an adaptive weight assignment mechanism to generate a rehabilitation start - goal semantic dynamic response encoding vector, including:

[0021] Calculate the decision - anchor adaptive splicing weight factors for each rehabilitation start - goal local semantic feature response anchor encoding matrix in the set of rehabilitation start - goal local semantic feature response anchor encoding matrices to obtain a set of rehabilitation start - goal decision - anchor adaptive splicing weight factors;

[0022] Fuse the set of rehabilitation start - goal local semantic feature response anchor encoding matrices based on the set of rehabilitation start - goal decision - anchor adaptive splicing weight factors to obtain a rehabilitation start - goal semantic dynamic response encoding vector.

[0023] Furthermore, calculate the decision - anchor adaptive splicing weight factors for each rehabilitation start - goal local semantic feature response anchor encoding matrix in the set of rehabilitation start - goal local semantic feature response anchor encoding matrices to obtain a set of rehabilitation start - goal decision - anchor adaptive splicing weight factors, including:

[0024] Based on the feature distributions of each rehabilitation start - goal local semantic feature response anchor encoding matrix in the set of rehabilitation start - goal local semantic feature response anchor encoding matrices, determine the rehabilitation start - goal decision - anchor adaptive splicing factors for each rehabilitation start - goal local semantic feature response anchor encoding matrix to obtain a set of rehabilitation start - goal decision - anchor adaptive splicing factors;

[0025] Perform weight - based processing on the set of rehabilitation start - goal decision - anchor adaptive splicing factors using the Softmax function to obtain a set of rehabilitation start - goal decision - anchor adaptive splicing weight factors.

[0026] Furthermore, based on the feature distributions of each rehabilitation start - goal local semantic feature response anchor encoding matrix in the set of rehabilitation start - goal local semantic feature response anchor encoding matrices, determine the rehabilitation start - goal decision - anchor adaptive splicing factors for each rehabilitation start - goal local semantic feature response anchor encoding matrix to obtain a set of rehabilitation start - goal decision - anchor adaptive splicing factors, including:

[0027] Based on the eigenvalues of each rehabilitation starting point - target local semantic feature response anchoring coding matrix in the set of rehabilitation starting point - target local semantic feature response anchoring coding matrices, a state transition relationship from the mean to the maximum value is constructed through the drift coefficient, and the intermediate transition weight global control is performed with the eigenvalue as the important score to obtain a set of adaptive splicing factors for the rehabilitation starting point - target decision anchor.

[0028] Furthermore, based on the rehabilitation starting point - target semantic dynamic response coding vector, a personalized rehabilitation plan is generated, including:

[0029] Input the rehabilitation starting point - target semantic dynamic response coding vector into a rehabilitation plan generator based on a large model to obtain a personalized rehabilitation plan.

[0030] According to another aspect of the present application, an intelligent analysis system for the rehabilitation treatment of lumbar disc herniation is provided, which includes:

[0031] A target patient object condition document acquisition module for acquiring the condition document of the target patient object, where the condition document includes a current condition part and a target part for the restoration of motor function requirements;

[0032] A semantic information extraction module for extracting the semantic information of the condition document to obtain a set of semantic coding vectors for the restoration of motor function requirements and a set of semantic coding vectors for the current condition part;

[0033] A rehabilitation starting point - target semantic analysis module for using the semantic coding vector for the restoration of motor function requirements as the rehabilitation target vector and the set of semantic coding vectors for the current condition part as the set of rehabilitation starting point vectors, and inputting them into a rehabilitation starting point - target semantic analysis network to obtain a rehabilitation starting point - target semantic dynamic response coding vector;

[0034] A personalized rehabilitation plan generation module for generating a personalized rehabilitation plan based on the rehabilitation starting point - target semantic dynamic response coding vector.

[0035] Compared with the prior art, the intelligent analysis system and method for the rehabilitation treatment of lumbar disc herniation provided by the present application convert abstract clinical information such as lesion site characteristics, pain perception descriptions, and motor function requirements into high - dimensional vectors by adopting a semantic coding technology based on deep learning, enabling the system to more accurately understand and simulate the patient's situation in a multi - dimensional space. Further, a dynamic mapping mechanism between the rehabilitation starting point and the target is established to capture the non - linear association between the patient's current state and the ideal rehabilitation target, and based on this, the intelligent derivation of the rehabilitation path is realized. Through the dynamic analysis of the patient's multi - dimensional information, the system can better understand the unique needs of each patient, improve the matching accuracy of the rehabilitation treatment plan, and thus make the treatment more personalized and effective. Description of the Drawings

[0036] The above and other objects, features, and advantages of the present application will become more apparent by describing the embodiments of the present application in more detail with reference to the accompanying drawings. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification, and are used to explain the present application together with the embodiments of the present application, and do not constitute a limitation to the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0037] Figure 1 is a flowchart of an intelligent analysis method for the rehabilitation treatment of lumbar disc herniation according to an embodiment of the present application;

[0038] Figure 2 is a schematic diagram of data flow of an intelligent analysis method for the rehabilitation treatment of lumbar disc herniation according to an embodiment of the present application;

[0039] Figure 3 is a flowchart of sub-step S3 of an intelligent analysis method for the rehabilitation treatment of lumbar disc herniation according to an embodiment of the present application;

[0040] Figure 4 is a flowchart of sub-step S32 of an intelligent analysis method for the rehabilitation treatment of lumbar disc herniation according to an embodiment of the present application;

[0041] Figure 5 is a block diagram of an intelligent analysis system for the rehabilitation treatment of lumbar disc herniation according to an embodiment of the present application. Detailed Embodiments

[0042] Next, exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. It should be understood that the present application is not limited by the exemplary embodiments described herein.

[0043] As shown in the present application and the claims, unless the context clearly indicates an exception, words such as "a", "an", "one", and / or "the" are not specifically singular and may also include the plural. Generally speaking, the terms "comprising" and "including" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list, and the method or device may also include other steps or elements.

[0044] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, however, any number of different modules can be used and run on the user terminal and / or the server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0045] In this application, flowcharts are used to illustrate the operations performed by the system according to the embodiments of this application. It should be understood that the operations before or below do not necessarily need to be executed precisely in sequence. Instead, according to requirements, various steps can be processed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0046] Next, exemplary embodiments according to this application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all the embodiments of this application. It should be understood that this application is not limited by the exemplary embodiments described herein.

[0047] In the technical solution of this application, an intelligent analysis method for the rehabilitation treatment of lumbar disc herniation is proposed. Figure 1 FIG. is a flowchart of an intelligent analysis method for the rehabilitation treatment of lumbar disc herniation according to an embodiment of this application. Figure 2 FIG. is a schematic diagram of data flow of an intelligent analysis method for the rehabilitation treatment of lumbar disc herniation according to an embodiment of this application. As Figure 1 and Figure 2 shown, the intelligent analysis method for the rehabilitation treatment of lumbar disc herniation according to the embodiment of this application includes the steps of: S1, obtaining the condition document of the target patient object, where the condition document includes a current condition part and a target part for the recovery of motor function; S2, extracting the semantic information of the condition document to obtain a set of semantic encoding vectors for the recovery of motor function and a set of semantic encoding vectors for the current condition part; S3, using the semantic encoding vector for the recovery of motor function as the rehabilitation target vector and the set of semantic encoding vectors for the current condition part as the set of rehabilitation starting point vectors, and inputting them into the rehabilitation starting point-target semantic analysis network to obtain a rehabilitation starting point-target semantic dynamic response encoding vector; S4, generating a personalized rehabilitation plan based on the rehabilitation starting point-target semantic dynamic response encoding vector.

[0048] In particular, in S1, obtain the medical condition document of the target patient object, where the medical condition document includes a current medical condition part and a target part for the recovery requirement of motor function. Among them, the current medical condition part includes basic patient information, descriptions related to the lesion site, and descriptions related to the pain level; the target part for the recovery requirement of motor function is the description related to the recovery requirement of motor function; among them, the basic patient information includes age, gender, weight, and medical history. It should be understood that the medical record document of each patient is unique, which contains rich unstructured text descriptions, such as the characteristics of the lesion site, the individualized expression of pain perception, etc., as well as structured data, such as age, gender, weight, and medical history. In the technical solution of the present application, by obtaining the medical condition document of the target patient object and extracting the basic patient information, descriptions related to the lesion site, descriptions related to the pain level, and descriptions related to the recovery requirement of motor function from the medical condition document of the target patient object, it can help the system understand the patient's health status and specific needs more comprehensively and accurately; in addition, by deeply analyzing the medical condition document, the specific situation of each patient can be accurately captured, including special needs and limiting conditions. In this way, the system can dynamically adjust the rehabilitation strategy according to the actual situation of the patient, which not only helps to improve the accuracy of diagnosis, but also provides a basis for formulating personalized treatment plans.

[0049] Specifically, in step S2, semantic information of the disease condition document is extracted to obtain a set of semantic encoding vectors for the motor function recovery requirement and a set of semantic encoding vectors for the current disease condition part. In a specific example of the present application, first, semantic embedding encoding is performed on descriptions related to the lesion site, descriptions related to the pain degree, descriptions related to the motor function recovery requirement, and the basic patient information to obtain a semantic encoding vector for the lesion site description, a semantic encoding vector for the pain degree description, a semantic encoding vector for the motor function recovery requirement, and a semantic encoding vector for the basic patient information. Considering that the patient's disease condition document contains a mixture of a large amount of unstructured medical descriptions (such as "tear of the annulus fibrosus at the L4-L5 segment with posterolateral protrusion of the nucleus pulposus") and structured physiological indicators (such as body mass index and duration of medical history), traditional methods use manual annotation or simple keyword matching to extract features, making it difficult to accurately capture complex semantics such as anatomical localization and pathological evolution stage implicit in the disease condition description. In the technical solution of the present application, descriptions related to the lesion site, descriptions related to the pain degree, descriptions related to the motor function recovery requirement, and the basic patient information are input into a semantic encoder based on the Bert model to obtain a semantic encoding vector for the lesion site description, a semantic encoding vector for the pain degree description, a semantic encoding vector for the motor function recovery requirement, and a semantic encoding vector for the basic patient information. Among them, Bert is a natural language processing model based on deep learning, and its core design concept is to achieve a deep understanding of text semantics through bidirectional context modeling and large-scale pre-training. In this process, the medical semantic understanding ability obtained by the Bert model through pre-training can automatically analyze the anatomical hierarchical relationship in the lesion site description and simultaneously capture the association pattern between subjective feelings and objective indicators in the pain degree description; for abstract goals such as motor function recovery requirements (such as "restore the ability to bend down and pick up objects"), the model extracts the implicit correspondence between the action function requirement and the spinal biomechanical constraint through the context attention mechanism; the basic patient information includes age, gender, weight, and medical history, and its individual physiological characteristics are integrated through semantic encoding, avoiding the problem of feature discretization caused by traditional numerical data processing. In this way, the system can not only automatically extract key information from unstructured disease condition descriptions but also comprehensively analyze in combination with the patient's personal information, significantly improving the system's ability to analyze the patient's disease condition and providing a solid foundation for achieving more scientific and personalized rehabilitation treatment. Furthermore, the semantic encoding vector for the lesion site description, the semantic encoding vector for the pain degree description, and the semantic encoding vector for the basic patient information are combined to obtain a set of semantic encoding vectors for the current disease condition part.

[0050] Specifically, in S3, taking the semantic encoding vector of the motor function recovery requirement as the rehabilitation target vector and the set of semantic encoding vectors of the current disease status as the set of rehabilitation starting vectors, inputting them into the rehabilitation starting point - target semantic analysis network to obtain the rehabilitation starting point - target semantic dynamic response encoding vector. It should be understood that there is a complex non - linear mapping relationship between the patient's current pathological state (such as L5 - S1 segment herniation with sciatica) and the expected rehabilitation goal (such as restoring lumbar forward flexion range of motion). Traditional methods use expert experience rules or linear regression models for path planning, and it is difficult to effectively model the dynamic coupling effect between the anatomical characteristics of the lesion site, the change of pain threshold, physiological feature constraints and the motor function recovery requirement. Therefore, in the technical solution of this application, taking the semantic encoding vector of the motor function recovery requirement as the rehabilitation target vector and the semantic encoding vectors of the lesion site description, pain degree description and patient basic information as the set of rehabilitation starting vectors, inputting them into the rehabilitation starting point - target semantic analysis network to obtain the rehabilitation starting point - target semantic dynamic response encoding vector. Specifically, this network extracts the deep implicit features of the rehabilitation target vector (i.e., the motor function recovery requirement) and the rehabilitation starting vectors (including the lesion site, pain degree and personal information), and constructs a local feature response matrix between the two based on the dynamic semantic matching mechanism, so as to generate a set of rehabilitation starting point - target local semantic feature response anchor encoding matrices. Then, through the adaptive weight allocation mechanism, multi - matrix weighted fusion is realized, and finally the rehabilitation starting point - target semantic dynamic response encoding vector is generated. That is, by deeply mining the potential association patterns between the set of starting vectors and the target vector, a dynamic response model with medical interpretability is constructed. In this way, not only the semantic fragmentation problem caused by the discrete distribution of multi - source feature vectors in the vector space is solved, but also a dynamic evolution path from the current state to the target state is established, overcoming the defect that the static scheme cannot adapt to the stage - by - stage changes such as pain relief and muscle function reconstruction in the rehabilitation process; in addition, through the dynamic response encoding mechanism, the influence weight of different rehabilitation starting points on the achievement of the target is quantified, avoiding the homogenization treatment of key decision - making factors in traditional methods. Furthermore, the system can identify and emphasize those factors that are most important for a specific patient or at a specific rehabilitation stage, making the generated rehabilitation plan more in line with the actual needs. Specifically, in a specific example of this application, as Figure 3 shown, S3 includes: S31, respectively extracting the deep implicit encoding features of the set of rehabilitation target vectors and the set of rehabilitation starting vectors, and constructing a local feature response matrix between the two based on the dynamic semantic matching mechanism to obtain a set of rehabilitation starting point - target local semantic feature response anchor encoding matrices; S32, realizing multi - matrix weighted fusion through the adaptive weight allocation mechanism to generate the rehabilitation starting point - target semantic dynamic response encoding vector.

[0051] Specifically, in S31, the deep implicit encoding features of the set of rehabilitation target vectors and the set of rehabilitation starting point vectors are respectively extracted, and a local feature response matrix between the two is constructed based on the dynamic semantic matching mechanism to obtain a set of rehabilitation starting point - target local semantic feature response anchor encoding matrices. In the embodiments of the present application, first, deep implicit feature extraction based on fully connected encoding is performed on the rehabilitation target vectors to obtain rehabilitation target feature deep implicit encoding vectors; that is, by using the semantic encoding vector of the motor function recovery requirement as the rehabilitation target vector and performing deep implicit feature extraction on it, it can be converted into a representation form with a higher dimension and rich semantic information. In this way, the system can understand the patient's rehabilitation needs at a higher dimension, thereby providing more accurate and personalized suggestions. In a specific example, the following feature extraction formula is used to perform deep implicit feature extraction based on fully connected encoding on the rehabilitation target vectors to obtain rehabilitation target feature deep implicit encoding vectors; where the feature extraction formula is:

[0052] ;

[0053] Among them, represents the rehabilitation target vector, is matrix multiplication, and are the rehabilitation target weight matrix and the rehabilitation target bias vector respectively, is the activation function, represents the rehabilitation target feature deep implicit encoding vector.

[0054] Next, perform deep implicit feature extraction based on fully connected encoding on each rehabilitation starting point vector in the set of rehabilitation starting point vectors to obtain a set of rehabilitation starting point feature deep implicit encoding vectors; by using a fully connected neural network to perform deep implicit feature extraction on the rehabilitation starting point vectors (i.e., the semantic encoding vectors of the lesion site description, the semantic encoding vectors of the pain degree description, and the semantic encoding vectors of the patient's basic information), these complex disease descriptions can be transformed into a higher-dimensional representation form with rich semantic information, thus better reflecting the actual situation of the patient. In this way, the system's ability to understand the patient's current health status can be enhanced, ensuring that the generated rehabilitation plan can accurately match the patient's needs. Specifically, by performing deep implicit feature extraction on each rehabilitation starting point vector, the system can learn a global representation of the patient's health status and capture the complex interaction relationships between the original features through a non-linear activation function. This deep information processing method helps to reveal the potential patterns behind the patient's condition and lays a foundation for the subsequent semantic alignment with the rehabilitation target vectors. In a specific example, perform deep implicit feature extraction based on fully connected encoding on each rehabilitation starting point vector in the set of rehabilitation starting point vectors according to the following feature extraction formula to obtain a set of rehabilitation starting point feature deep implicit encoding vectors; where the feature extraction formula is:

[0055] ;

[0056] ;

[0057] Where represents the set of rehabilitation starting point vectors, , , and respectively represent the 1st, 2nd, rd, and th rehabilitation starting point vectors in the set of rehabilitation starting point vectors, and are the rehabilitation starting point weight matrix and the rehabilitation starting point bias vector respectively, represents the rd rehabilitation starting point feature deep implicit encoding vector in the set of rehabilitation starting point feature deep implicit encoding vectors.

[0058] Furthermore, perform start - goal semantic dynamic response encoding on each start - point feature deep implicit encoding vector in the set of the rehabilitation goal feature deep implicit encoding vector and the start - point feature deep implicit encoding vector to obtain a set of start - goal local semantic feature response anchor encoding matrices. Considering that traditional rehabilitation assessment methods often rely on doctors' experience and static indicators, it is difficult for this method to comprehensively consider the individual differences and complex conditions of patients. For example, different patients may exhibit different symptoms, pain sensations, and recovery needs at the same disease stage, and these factors will all affect the final rehabilitation effect. By performing start - goal semantic dynamic response encoding on each start - point feature deep implicit encoding vector in the set of the rehabilitation goal feature deep implicit encoding vector and the start - point feature deep implicit encoding vector, the system can identify the complex interactions between these features and their impacts on the rehabilitation process. Specifically, the generated set of start - goal local semantic feature response anchor encoding matrices not only represents the interaction information between the rehabilitation goal and the start - point features but also serves as a bridge for capturing the global and local semantic alignment relationships. By constructing the local feature response anchor encoding matrix, the system can understand the patient's health status in a higher dimension and identify the details that are crucial for the rehabilitation process. In a specific example, perform start - goal semantic dynamic response encoding on each start - point feature deep implicit encoding vector in the set of the rehabilitation goal feature deep implicit encoding vector and the start - point feature deep implicit encoding vector with the following dynamic response encoding formula to obtain a set of start - goal local semantic feature response anchor encoding matrices; where the dynamic response encoding formula is:

[0059] ;

[0060] where, is the transposed vector of is the length of denotes vector multiplication, denotes the th start - goal local semantic feature response anchor encoding matrix in the set of start - goal local semantic feature response anchor encoding matrices.

[0061] Specifically, in S32, implement multi - matrix weighted fusion through an adaptive weight assignment mechanism to generate a start - goal semantic dynamic response encoding vector. In a specific example of the present application, as Figure 4As shown, the S32 includes: S321, calculating the decision anchor adaptive splicing weight factors of each rehabilitation starting point-target local semantic feature response anchor encoding matrix in the set of rehabilitation starting point-target local semantic feature response anchor encoding matrices to obtain a set of rehabilitation starting point-target decision anchor adaptive splicing weight factors; S322, fusing the set of rehabilitation starting point-target local semantic feature response anchor encoding matrices based on the set of rehabilitation starting point-target decision anchor adaptive splicing weight factors to obtain a rehabilitation starting point-target semantic dynamic response encoding vector.

[0062] More specifically, in S321, the decision anchor adaptive splicing weight factors of each rehabilitation starting point - target local semantic feature response anchor encoding matrix in the set of rehabilitation starting point - target local semantic feature response anchor encoding matrices are calculated to obtain a set of rehabilitation starting point - target decision anchor adaptive splicing weight factors. In the technical solution of this application, first, based on the eigenvalues of each rehabilitation starting point - target local semantic feature response anchor encoding matrix in the set of rehabilitation starting point - target local semantic feature response anchor encoding matrices, a state transition relationship from the mean to the maximum value is constructed through a drift coefficient, and the global control of the intermediate transition weight is performed with the eigenvalue as the importance score to obtain a set of rehabilitation starting point - target decision anchor adaptive splicing factors; that is, by analyzing the feature distribution of the rehabilitation starting point - target local semantic feature response anchor encoding matrix, the rehabilitation starting point - target decision anchor adaptive splicing factor is determined. In this way, the system can identify the dominance of each local feature in the rehabilitation task, and based on this, the system can dynamically adjust the importance of features at both the global and local levels, so as to ensure that the generated rehabilitation plan can accurately match the actual situation of the patient. For example, if the main problem of a certain patient is severe pain rather than limited motor function, then the system will correspondingly increase the weight of pain management and reduce the proportion of exercise training. In particular, it should be understood that the feature importance in different rehabilitation stages of patients shows a non-uniform distribution characteristic. For example, the pain degree feature may dominate the treatment priority in the acute inflammation reaction stage, while the weight of the motor function demand feature significantly increases in the chronic recovery period. In the technical solution of this application, based on the eigenvalues of each rehabilitation starting point - target local semantic feature response anchor encoding matrix in the set of rehabilitation starting point - target local semantic feature response anchor encoding matrices, a state transition relationship from the mean to the maximum value is constructed through a drift coefficient, and the global control of the intermediate transition weight is performed with the eigenvalue as the importance score to obtain a set of rehabilitation starting point - target decision anchor adaptive splicing factors. Specifically, by introducing a state transition mechanism driven by a drift coefficient, it aims to establish a dynamically interpretable generalization path from mean statistics (weakly interpretable overall distribution) to maximum value focus (strongly interpretable significant features). That is, a dynamic balance mechanism is established between the statistical characteristics of the feature distribution (the overall trend represented by the mean) and the pathological significance (the emergency state pointed to by the maximum value), simulating the comprehensive weighing process of clinical experts for "the current most urgent problem" and "long-term risk control" during the formulation of the plan. Specifically, in this process, the eigenvalue participates in the global control of the intermediate transition weight as the importance score. When formulating a rehabilitation plan, it is necessary to comprehensively consider the multi-dimensional information of the patient (such as the mean background of age, medical history, etc.), and at the same time, the decision weight of key pathological indicators (such as severe pain or specific intervertebral disc compression and other local strong features) will be emphasized.For example, when strong semantic features such as "radiating severe pain" frequently appear in the patient's pain description, the system enhances the weight of the corresponding local response matrix through the eigenvalue importance score, so that the rehabilitation plan gives priority to analgesic intervention measures rather than conventional exercise training. This global weight control based on state transition enables the system to simulate the progressive adjustment logic of the clinical rehabilitation path and achieve truly personalized dynamic plan optimization while ensuring medical standardization. In a specific example, based on the feature distribution of each rehabilitation starting point-goal local semantic feature response anchoring coding matrix in the set of rehabilitation starting point-goal local semantic feature response anchoring coding matrices, the rehabilitation starting point-goal decision anchor adaptive splicing factor of each rehabilitation starting point-goal local semantic feature response anchoring coding matrix is determined by the following decision anchor adaptive splicing factor calculation formula to obtain a set of rehabilitation starting point-goal decision anchor adaptive splicing factors; wherein, the decision anchor adaptive splicing factor calculation formula is:.

[0063] ;

[0064] ;

[0065] ;

[0066] ;

[0067] Wherein, represents the variance of, represents the mean of, represents the number of eigenvalues of as the difference amplification coefficient, is for calculating the number of eigenvalues of, represents the maximum value of ， represents the drift coefficient, is the intermediate state transition, is the th eigenvalue of, represents the exponential function with the natural constant as the base, represents the th rehabilitation starting point-goal decision anchor adaptive splicing factor in the set of rehabilitation starting point-goal decision anchor adaptive splicing factors.

[0068] Next, perform a weighting process based on the Softmax function on the set of rehabilitation starting point - target decision anchor adaptive splicing factors to obtain a set of rehabilitation starting point - target decision anchor adaptive splicing weight factors. It should be understood that the dynamic association between the patient's rehabilitation starting point features (such as lesion site description, pain perception, individual basic information) and the rehabilitation goal (motor function recovery needs) has significant non - linearity and uncertainty. Traditional fixed - weight allocation methods are difficult to adapt to the complexity hidden in the data distributions of different patients. If directly using the original splicing factors for feature fusion, due to the lack of a normalization quantization standard for the contribution degrees of each local response matrix, key features will be weakened or redundant features will be over - amplified. In the technical solution of this application, a weighting process based on the Softmax function is performed on the set of rehabilitation starting point - target decision anchor adaptive splicing factors to obtain a set of rehabilitation starting point - target decision anchor adaptive splicing weight factors. By using the Softmax function, the system can convert the original decision anchor adaptive splicing factors into an easily interpretable weight distribution. In this way, the system can dynamically adjust its weights according to the specific importance of each local feature, so that those features crucial to the rehabilitation process are given higher priorities. For example, in a certain patient's rehabilitation plan, if pain management is more critical than motor function recovery, then the corresponding pain - related features will occupy a larger proportion in the final rehabilitation plan. In a specific example, the following weight calculation formula is used to perform a weighting process based on the Softmax function on the set of rehabilitation starting point - target decision anchor adaptive splicing factors to obtain a set of rehabilitation starting point - target decision anchor adaptive splicing weight factors; where the weight calculation formula is:

[0069] ;

[0070] where, is a normalization function, represents the th rehabilitation starting point - target decision anchor adaptive splicing weight factor in the set of rehabilitation starting point - target decision anchor adaptive splicing weight factors.

[0071] More specifically, in S322, the set of local semantic feature response anchor encoding matrices for the rehabilitation start - goal is fused based on the set of adaptive splicing weight factors for the rehabilitation start - goal decision anchor to obtain the semantic dynamic response encoding vector for the rehabilitation start - goal. That is, by fusing each local semantic feature response anchor encoding matrix according to its corresponding adaptive splicing weight factor, the system can comprehensively consider all relevant factors at the global level to form a dynamic response encoding vector that comprehensively reflects the patient's health status. This approach enables the system to flexibly adjust the rehabilitation strategy according to the specific situation of each patient. For example, if the main problem of a certain patient is severe pain rather than limited motor function, then the system will correspondingly increase the weight of pain management and reduce the proportion of physical training, thus generating a personalized rehabilitation plan more suitable for that patient. In a specific example, the set of local semantic feature response anchor encoding matrices for the rehabilitation start - goal is fused based on the following fusion formula with the set of adaptive splicing weight factors for the rehabilitation start - goal decision anchor to obtain the semantic dynamic response encoding vector for the rehabilitation start - goal; where the fusion formula is:

[0072] ;

[0073] where is the dynamic response encoding matrix for the rehabilitation start - goal, is the shape reshaping operation, represents the semantic dynamic response encoding vector for the rehabilitation start - goal.

[0074] Specifically, in S4, a personalized rehabilitation plan is generated based on the rehabilitation starting point-goal semantic dynamic response encoding vector. In the technical solution of this application, the rehabilitation starting point-goal semantic dynamic response encoding vector is input into a rehabilitation plan generator based on a large model to obtain a personalized rehabilitation plan. It should be understood that the multi-dimensional feature vectors (such as lesion site features, pain dynamic evolution patterns, individualized rehabilitation goals) after semantic analysis and dynamic response encoding are essentially abstract representations in a high-dimensional hidden space, and the mapping relationship between them and specific treatment means (such as physical therapy intensity, exercise training type, drug intervention strategy) has highly non-linear and cross-modal correlation characteristics. Existing rule engines or simple classification models are difficult to capture this implicit association from complex semantic features to multi-dimensional rehabilitation actions. In the technical solution of this application, by inputting the rehabilitation starting point-goal semantic dynamic response encoding vector into a rehabilitation plan generator based on a large model, an end-to-end mapping reconstruction from the semantic feature space to a clinically executable plan is realized, that is, through the pre-trained understanding of a large model of massive cross-modal medical knowledge (such as clinical guidelines, rehabilitation medicine literature, patient treatment trajectory data), the individualized rehabilitation needs (such as acute-phase analgesia priority, chronic-phase functional remodeling) implicit in the dynamic response encoding vector are transformed into specific treatment steps with temporal logic and dose control. In this way, the system can more deeply understand the complex interaction between each local feature response anchoring encoding matrix, thereby generating more accurate rehabilitation suggestions.

[0075] In summary, the intelligent analysis method for the rehabilitation treatment of lumbar disc herniation according to the embodiments of this application is clarified. By using a semantic encoding technology based on deep learning to transform abstract clinical information such as lesion site features, pain perception descriptions, and motor function requirements into high-dimensional vectors, the system can more accurately understand and simulate the patient's situation in a multi-dimensional space. Further, a dynamic mapping mechanism between the rehabilitation starting point and the goal is established to capture the non-linear association between the patient's current state and the ideal rehabilitation goal, and based on this, the intelligent derivation of the rehabilitation path is realized. Through the dynamic analysis of the patient's multi-dimensional information, the system can better understand the unique needs of each patient, improve the matching accuracy of the rehabilitation treatment plan, and thus make the treatment more personalized and effective.

[0076] Furthermore, an intelligent analysis system for the rehabilitation treatment of lumbar disc herniation is also provided.

[0077] Figure 5 FIG. is a block diagram of the intelligent analysis system for the rehabilitation treatment of lumbar disc herniation according to the embodiments of this application. As Figure 5As shown, an intelligent analysis system 300 for the rehabilitation treatment of lumbar disc herniation according to an embodiment of the present application includes: a target patient object condition document acquisition module 310, configured to acquire the condition document of the target patient object, where the condition document includes a current condition part and a target part for the recovery of motor function; a semantic information extraction module 320, configured to extract the semantic information of the condition document to obtain a set of semantic encoding vectors for the recovery of motor function and semantic encoding vectors for the current condition part; a rehabilitation starting point - target semantic analysis module 330, configured to use the semantic encoding vector for the recovery of motor function as the rehabilitation target vector and the set of semantic encoding vectors for the current condition part as the set of rehabilitation starting point vectors, and input them into a rehabilitation starting point - target semantic analysis network to obtain a rehabilitation starting point - target semantic dynamic response encoding vector; and a personalized rehabilitation plan generation module 340, configured to generate a personalized rehabilitation plan based on the rehabilitation starting point - target semantic dynamic response encoding vector.

[0078] As described above, the intelligent analysis system 300 for the rehabilitation treatment of lumbar disc herniation according to an embodiment of the present application can be implemented in various wireless terminals, such as a server having an intelligent analysis algorithm for the rehabilitation treatment of lumbar disc herniation. In a possible implementation manner, the intelligent analysis system 300 for the rehabilitation treatment of lumbar disc herniation according to an embodiment of the present application can be integrated into a wireless terminal as a software module and / or a hardware module. For example, the intelligent analysis system 300 for the rehabilitation treatment of lumbar disc herniation can be a software module in the operating system of the wireless terminal, or can be an application program developed for the wireless terminal; of course, the intelligent analysis system 300 for the rehabilitation treatment of lumbar disc herniation can also be one of the many hardware modules of the wireless terminal.

[0079] Alternatively, in another example, the intelligent analysis system 300 for the rehabilitation treatment of lumbar disc herniation and the wireless terminal can also be separate devices, and the intelligent analysis system 300 for the rehabilitation treatment of lumbar disc herniation can be connected to the wireless terminal through a wired and / or wireless network, and transmit and interact information in accordance with a predefined data format.

[0080] The various embodiments of the present disclosure have been described above. The above description is exemplary and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations are obvious to those of ordinary skill in the art in the technical field without departing from the scope and spirit of the described embodiments. The selection of the terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other ordinary skilled persons in the technical field to understand the embodiments disclosed herein.

Claims

1. An intelligent analysis method for the rehabilitation treatment of lumbar disc herniation, characterized in that, Including: Obtain the medical condition document of the target patient object, where the medical condition document includes a current medical condition part and a target part for the recovery requirement of motor function; Extract the semantic information of the medical condition document to obtain a set of semantic encoding vectors for the recovery requirement of motor function and semantic encoding vectors for the current medical condition part; Using the semantic encoding vector for the recovery requirement of motor function as the rehabilitation target vector and the set of semantic encoding vectors for the current medical condition part as the set of rehabilitation starting point vectors, input them into the rehabilitation starting point - target semantic analysis network to obtain the rehabilitation starting point - target semantic dynamic response encoding vector, including: respectively extract the deep implicit encoding features of the rehabilitation target vector and the set of rehabilitation starting point vectors, and construct a local feature response matrix between them based on the dynamic semantic matching mechanism to obtain a set of rehabilitation starting point - target local semantic feature response anchored encoding matrices; realize multi - matrix weighted fusion through the adaptive weight allocation mechanism to generate the rehabilitation starting point - target semantic dynamic response encoding vector; Generate a personalized rehabilitation plan based on the rehabilitation starting point - target semantic dynamic response encoding vector; Among them, respectively extract the deep implicit encoding features of the rehabilitation target vector and the set of rehabilitation starting point vectors, and construct a local feature response matrix between them based on the dynamic semantic matching mechanism to obtain a set of rehabilitation starting point - target local semantic feature response anchored encoding matrices, including: Perform deep implicit feature extraction based on fully - connected encoding on the rehabilitation target vector to obtain the rehabilitation target feature deep implicit encoding vector; Perform deep implicit feature extraction based on fully - connected encoding on each rehabilitation starting point vector in the set of rehabilitation starting point vectors to obtain a set of rehabilitation starting point feature deep implicit encoding vectors; Perform rehabilitation starting point - target semantic dynamic response encoding on the rehabilitation target feature deep implicit encoding vector and each rehabilitation starting point feature deep implicit encoding vector in the set of rehabilitation starting point feature deep implicit encoding vectors to obtain a set of rehabilitation starting point - target local semantic feature response anchored encoding matrices.

2. The intelligent analysis method for rehabilitation treatment of lumbar disc herniation according to claim 1, wherein The current medical condition part includes the patient's basic information, descriptions related to the lesion site, and descriptions related to the pain level; The target part for the recovery requirement of motor function is the description related to the recovery requirement of motor function.

3. The intelligent analysis method for rehabilitation treatment of lumbar disc herniation according to claim 2, wherein, Extract the semantic information of the medical condition document to obtain a set of semantic encoding vectors for the recovery requirement of motor function and semantic encoding vectors for the current medical condition part, including: Perform semantic embedding encoding on the descriptions related to the lesion site, descriptions related to the pain level, descriptions related to the recovery requirement of motor function, and the patient's basic information to obtain the semantic encoding vector for the lesion site description, the semantic encoding vector for the pain level description, the semantic encoding vector for the recovery requirement of motor function, and the semantic encoding vector for the patient's basic information; Merge the semantic encoding vector for the lesion site description, the semantic encoding vector for the pain level description, and the semantic encoding vector for the patient's basic information to obtain a set of semantic encoding vectors for the current medical condition part.

4. The intelligent analysis method for rehabilitation treatment of lumbar disc herniation according to claim 2, wherein, The patient's basic information includes age, gender, weight, and medical history.

5. The intelligent analysis method for the rehabilitation treatment of lumbar disc herniation according to claim 1, characterized in that, Realize multi - matrix weighted fusion through the adaptive weight allocation mechanism to generate the rehabilitation starting point - target semantic dynamic response encoding vector, including: Calculate the decision anchor adaptive splicing weight factors for each rehabilitation starting point - target local semantic feature response anchor encoding matrix in the set of rehabilitation starting point - target local semantic feature response anchor encoding matrices to obtain the set of rehabilitation starting point - target decision anchor adaptive splicing weight factors; Based on the set of rehabilitation starting point - target decision anchor adaptive splicing weight factors, fuse the set of rehabilitation starting point - target local semantic feature response anchor encoding matrices to obtain the rehabilitation starting point - target semantic dynamic response encoding vector.

6. The intelligent analysis method for rehabilitation treatment of lumbar disc herniation according to claim 5, wherein Calculate the decision anchor adaptive splicing weight factors for each rehabilitation starting point - target local semantic feature response anchor encoding matrix in the set of rehabilitation starting point - target local semantic feature response anchor encoding matrices to obtain the set of rehabilitation starting point - target decision anchor adaptive splicing weight factors, including: Based on the feature distributions of each rehabilitation starting point - target local semantic feature response anchor encoding matrix in the set of rehabilitation starting point - target local semantic feature response anchor encoding matrices, determine the rehabilitation starting point - target decision anchor adaptive splicing factors for each rehabilitation starting point - target local semantic feature response anchor encoding matrix to obtain the set of rehabilitation starting point - target decision anchor adaptive splicing factors; Perform weight processing based on the Softmax function on the set of rehabilitation starting point - target decision anchor adaptive splicing factors to obtain the set of rehabilitation starting point - target decision anchor adaptive splicing weight factors.

7. The intelligent analysis method for the rehabilitation treatment of lumbar disc herniation according to claim 6, characterized in that, Based on the feature distributions of each rehabilitation starting point - target local semantic feature response anchor encoding matrix in the set of rehabilitation starting point - target local semantic feature response anchor encoding matrices, determine the rehabilitation starting point - target decision anchor adaptive splicing factors for each rehabilitation starting point - target local semantic feature response anchor encoding matrix to obtain the set of rehabilitation starting point - target decision anchor adaptive splicing factors, including: Based on the eigenvalues of each rehabilitation starting point - target local semantic feature response anchor encoding matrix in the set of rehabilitation starting point - target local semantic feature response anchor encoding matrices, construct a state transition relationship from the mean to the maximum value through the drift coefficient, and globally control the intermediate transition weights with the eigenvalues as the importance scores to obtain the set of rehabilitation starting point - target decision anchor adaptive splicing factors.

8. The intelligent analysis method for rehabilitation treatment of lumbar disc herniation according to claim 1, characterized in that Based on the rehabilitation starting point - target semantic dynamic response encoding vector, generate a personalized rehabilitation plan, including: Input the rehabilitation starting point - target semantic dynamic response encoding vector into a rehabilitation plan generator based on a large model to obtain a personalized rehabilitation plan.

9. An intelligent analysis system for the rehabilitation treatment of lumbar disc herniation, using the intelligent analysis method for the rehabilitation treatment of lumbar disc herniation described in claim 1, characterized in that, Including: A target patient object condition document acquisition module for acquiring the condition document of the target patient object, where the condition document includes a current condition part and a target part for the recovery of motor function; A semantic information extraction module for extracting the semantic information of the condition document to obtain a set of semantic encoding vectors for the recovery of motor function and the current condition part; A rehabilitation starting point - target semantic analysis module for using the semantic encoding vector for the recovery of motor function as the rehabilitation target vector and the set of semantic encoding vectors of the current condition part as the set of rehabilitation starting point vectors, and inputting them into the rehabilitation starting point - target semantic analysis network to obtain the rehabilitation starting point - target semantic dynamic response encoding vector; A personalized rehabilitation plan generation module, which is used to generate a personalized rehabilitation plan based on the rehabilitation starting point-goal semantic dynamic response encoding vector.

Images