SOAP document generation method based on operation type staging and multi-source data linkage
By using the SOAP document generation method, which links surgical procedures with multi-source data, the problem of insufficient data interoperability in rehabilitation documentation systems has been solved. This has enabled automated and intelligent document generation, improved efficiency and the objectivity of assessment conclusions, and reduced medical risks.
Patent Information
- Application Number
- CN202611379164.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-07
- Publication Date
- 2026-11-03
AI Technical Summary
Existing rehabilitation documentation systems lack data sharing mechanisms, requiring therapists to manually input data, which is inefficient. Furthermore, assessment conclusions are highly subjective, prone to data mismatch, and may lead to discrepancies between plans and reality, posing medical risks. The documentation also lacks structured feedback and cannot be used for quality statistics and intelligent decision-making.
The SOAP document generation method, based on surgical procedure staging and multi-source data linkage, is adopted. Through a specialist rule engine and structured knowledge graph, it achieves automatic data alignment and filling, automatically generates structured evaluation analysis, and sets up triple logical checks to ensure the integrity and consistency of the documents.
It has achieved full automation and intelligent generation of rehabilitation documents, improving efficiency, eliminating data mismatch and medical risks, ensuring the objectivity of assessment conclusions, and supporting scientific research analysis and quality statistics.
Smart Images
Figure CN123033644A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of document generation technology, specifically to a SOAP document generation method based on technique-based staging and multi-source data linkage. Background Technology
[0002] SOAP (Subjective, Objective, Assessment, Plan) is an internationally recognized format for rehabilitation documentation, widely recommended by the World Federation of Physical Therapy (WCPT) and the American Academy of Physical Therapy (APTA). Currently, while most medical institutions in China have deployed electronic medical record systems for rehabilitation documentation, their application remains at the stage of digitizing paper documents—merely converting document format—far from achieving intelligent and automated levels.
[0003] Existing rehabilitation documentation systems lack effective data exchange mechanisms with various hospital information systems, forcing therapists to manually record information for each document. Clinical statistics show that therapists spend over 30% of their daily work time writing assessment, treatment, and progress notes. A complete SOAP document takes an average of 30 minutes, requiring repeated queries, copying, and pasting of data from multiple isolated systems such as HIS, LIS, PACS, and rehabilitation assessment terminals, resulting in extremely low efficiency.
[0004] At the same time, because rehabilitation-related data such as assessment data, prescription information, execution records, test indicators and imaging measurements are stored in separate heterogeneous systems, and there is a lack of automated data association and time alignment mechanisms between the systems, therapists need to associate and align these scattered data one by one when integrating them. This process is not only inefficient, but also prone to problems such as data mismatch, unit confusion and omission of key information, leading to inaccurate assessment conclusions.
[0005] In the assessment and analysis (A) stage, this part of the documentation relies entirely on the therapist's personal clinical experience and subjective judgment. The system itself cannot automatically identify the patient's surgical procedure and current postoperative stage, nor can it cite the rehabilitation milestones corresponding to that stage. As a result, the analysis results are highly subjective, vary greatly, and cannot objectively measure the patient's functional recovery level.
[0006] In the treatment plan (P) stage, the treatment plan recorded in this part is generally inconsistent with the actual effective rehabilitation prescription. When the therapist adjusts the training intensity or contraindicated movements in the prescription system, the therapist often does not update the corresponding plan in SOAP at the same time. The existing system does not have any consistency verification or risk warning, which leads to the discrepancy between the plan and the actual situation, and is prone to medical disputes and safety hazards.
[0007] Furthermore, current SOAP documents are generally stored in an unstructured, free-text format, and their content cannot be parsed by machines. Assessment conclusions and plan adjustment records are buried in the text and cannot be structurally fed back to the rehabilitation prescription system to trigger stage switching or prescription evolution. They also cannot be used for quality statistics, multi-center scientific research analysis, or intelligent decision-making model training. Summary of the Invention
[0008] The present invention aims to provide a SOAP document generation method based on surgical stage and multi-source data linkage, which solves the problems of time-consuming, non-standard, incomplete, data silos and medical risks in existing SOAP document generation.
[0009] To achieve the above objectives, the present invention adopts the following technical solution: a SOAP document generation method based on surgical procedure staging and multi-source data linkage, comprising the following steps: S1. Obtain patient identity information and rehabilitation scenario, and dynamically determine the patient's surgical procedure and current rehabilitation stage by combining multi-dimensional data; match and retrieve the corresponding SOAP template from the pre-set structured document template library, wherein the SOAP template has 4 types of specialized feature chapters; S2 retrieves multi-source heterogeneous data, filters and evaluates the latest valid data within the time window using the time alignment engine, and automatically fills it into the objective data section; it also retrieves the current patient's execution completion data and fills it into the treatment plan and evaluation analysis sections. S3 retrieves the surgical type-stage quantitative milestone target database, compares objective data, and automatically generates a structured evaluation and analysis draft; S4 integrates the data generated above in SOAP format, performs triple logical checks, automatically marks abnormal items and forces corrections, generates the final SOAP document, and synchronizes it to the HIS system for archiving after confirmation.
[0010] The principle of this scheme is: This solution, based on a three-layer technical architecture of specialized rule engine, multi-source data linkage, and structured knowledge graph, breaks through the technical bottleneck of traditional document systems that only support manual input and simple template replacement, achieving fully automated and intelligent generation of SOAP rehabilitation documents. Its technical principle lies in: 1. Dynamic Dual-Dimensional Determination of Surgical Procedure and Stage: The system has a built-in knowledge graph of staging rules specific to each surgical procedure, covering the four-stage rehabilitation assessment criteria for 12 common orthopedic surgical procedures. Based on the patient's surgical procedure information, five types of dynamic data—postoperative days, wound healing grade, joint range of motion, muscle strength grade, and weight-bearing status—are used as input parameters. The deterministic finite automaton (DFA) algorithm is used for rule matching to dynamically determine the patient's current rehabilitation stage, completely replacing manual staging assessment.
[0011] 2. Time Alignment and Fusion of Multi-Source Heterogeneous Medical Data: By connecting to 8 types of heterogeneous medical systems, the system uses a timestamp master index to achieve automatic alignment of cross-system data. Each piece of data is automatically labeled with a collection timestamp. Based on the document evaluation time, overdue data that exceeds the time window is filtered out. For multiple records of the same indicator, a weighted algorithm of nearest time priority + clinical priority is used to select effective data. Finally, a rehabilitation data view with the patient as the center and a unified time dimension is formed, which completely solves the industry problems of fragmentation and time mismatch of multi-source data.
[0012] 3. Automatic generation of structured assessment and analysis: This solution has a built-in quantitative rehabilitation milestone target library for specific surgical procedures and stages. Through a vector similarity matching algorithm, the objective data of patients are compared with the milestone targets of the corresponding stages one by one, and the "achieved / not achieved" judgment results are automatically generated. The solution also automatically generates the analysis of the reasons for not achieving the target and suggestions for improvement of rehabilitation training through preset clinical reasoning rules, so as to achieve a completely data-driven assessment and analysis and eliminate the subjective differences of manual analysis.
[0013] 4. Cross-module logic verification and closed-loop feedback: This solution designs a three-layer verification mechanism involving data, logic, and business. The data layer verifies the rationality of numerical values and the completeness of fields; the logic layer verifies the internal consistency of the four parts (S / O / A / P); and the business layer verifies the consistency between the treatment plan and the current valid prescription. Anomalies are automatically highlighted in red using regular expression matching and trigger mandatory correction reminders. The generated structured assessment conclusions are pushed to the rehabilitation prescription system in real time via a RESTful interface, triggering rehabilitation stage switching or prescription parameter optimization, forming a complete business closed loop of "assessment-prescription-execution-reassessment," and realizing intelligent management of the entire rehabilitation process.
[0014] The advantages of this solution are: 1. The SOAP template adaptive matching mechanism set up by this solution, driven by both surgical type and stage, is different from general templates. It can accurately match specialized and staged SOAP document templates according to the patient's specific surgical method and dynamically determined rehabilitation stage, which is the first of its kind in the industry.
[0015] 2. This solution breaks down multiple data silos within and outside the hospital (HIS, LIS, PACS, rehabilitation equipment, wearable devices), enabling automatic retrieval of multi-source heterogeneous data, time alignment, and fully automatic filling of objective data (O). The time alignment engine ensures data timeliness and achieves zero manual entry of objective data sections, rather than simple text replacement.
[0016] 3. For the first time, the generation of Part A is compared in a structured manner with the surgical type-stage quantitative rehabilitation goals, automatically generating data-driven, objective and quantifiable assessment and analysis conclusions, completely changing the traditional subjective writing mode that relies on personal experience.
[0017] 4. It pioneered a real-time comparison and warning mechanism between the P (Plan) component and the actual effective prescriptions, solving the problem of complete separation between clinical intractable disease plans and prescriptions. At the same time, it provides structured feedback of SOAP assessment results to the prescription system, driving stage switching and prescription evolution, forming a complete safety closed loop of "assessment-prescription-execution-reassessment".
[0018] 5. Automatically generates standard SOAP documents and enforces triple checks for integrity, numerical range, and cross-part logical consistency. Existing technologies and systems do not possess such rigorous automated quality control capabilities. Attached Figure Description
[0019] Figure 1 This is a flowchart illustrating the SOAP document generation method based on surgical procedure staging and multi-source data linkage of the present invention. Detailed Implementation
[0020] The following detailed description illustrates the specific implementation method: The SOAP document generation method based on surgical stage and multi-source data linkage in this embodiment connects HIS, LIS, PACS, rehabilitation assessment equipment, wearable devices and prescription systems through standardized data interfaces. It realizes a closed loop of the entire process, including assessment data, prescription execution data, automatic document filling, SOAP structure assembly, logical verification, electronic signature, and archiving synchronization. It automatically generates standardized rehabilitation documents that meet the requirements of tertiary hospital review and medical insurance standards, and completely solves the core pain points such as time-consuming, non-standard, incomplete, data silos and medical risks.
[0021] A SOAP document generation method based on surgical procedure staging and multi-source data linkage is provided, as shown in the appendix. Figure 1 As shown, it includes the following steps: S1: Obtain patient identity information and rehabilitation scenario, and dynamically determine the patient's surgical procedure and current rehabilitation stage by combining multi-dimensional data; match and retrieve the corresponding SOAP template from the pre-set structured document template library. The SOAP template has 4 types of specialized feature chapters.
[0022] In this embodiment, patient identity information (name, medical record number) is obtained and automatically extracted from HIS surgical records or EMR or entered by the doctor: surgical procedure name (e.g., "arthroscopic anterior cruciate ligament reconstruction"), surgical date, and surgical method (arthroscopic / open surgery).
[0023] Subsequently, the automatic determination module for surgical type and stage (which can independently implement the exercise prescription system) is invoked to dynamically determine the patient's current rehabilitation stage based on the number of days after surgery, wound healing status, key assessment data (such as joint range of motion, muscle strength, and weight-bearing status) and the milestone target library.
[0024] Specifically, the dynamic determination steps for rehabilitation staging are as follows: The system calls upon a pre-defined staging rule library specific to each surgical procedure, taking the patient's postoperative days, wound healing level, active joint range of motion measurement, manual muscle strength assessment, and weight-bearing capacity as input parameters. Through rule matching, it automatically determines the current rehabilitation stage, which includes the healing and repair stage, early rehabilitation stage, strength and control stage, and return to exercise stage.
[0025] Based on the identified surgical procedure and rehabilitation stage, the system automatically matches and retrieves the corresponding SOAP format document template from a pre-set structured document template library. In this embodiment, the four types of specialized feature sections include: Subjective data section: defined as S, pre-sets inquiry points related to the surgical procedure and stage; such as "whether there is a popping / locking sensation in the knee joint after surgery" and "whether night pain affects sleep".
[0026] Objective data section: Defined as O, it includes a list of objective indicators that must be evaluated in each stage; for example, the healing and repair stage must include "wound condition, swelling value, and active knee flexion angle".
[0027] The assessment and analysis section, defined as A, embeds a quantitative milestone target library for the surgical procedure and stage, which is constructed from the AOPT, AAOS guidelines, and expert consensus.
[0028] Treatment plan section: Defined as P, with reserved data fields for linkage with the prescription system.
[0029] S2 retrieves multi-source heterogeneous data, filters and evaluates the latest valid data within the time window using the time alignment engine, and automatically fills it into the objective data section; it also retrieves the current patient's execution completion data and fills it into the treatment plan section and the evaluation analysis section.
[0030] In this embodiment, the system automatically pulls the latest data from various data sources through a standardized interface, processes it through a time alignment engine, and then automatically fills the corresponding fields in Chapter O.
[0031] The multi-source heterogeneous data includes: surgical records, vital signs (heart rate, blood pressure), and comorbidity information from the HIS / EMR system; laboratory indicators such as CRP, blood routine, and inflammatory factors from the LIS system; postoperative joint alignment angles and implant position image measurements from the PACS system; joint range of motion, muscle strength grading (MMT), balance score (Berg), and 6-minute walking distance data from rehabilitation assessment equipment terminals; step count, activity intensity, and sleep duration data from wearable devices; training exercise completion count, number of sets, completion rate, and abnormal reaction records from the prescription execution system; and home self-test data such as active joint range of motion and single-leg standing time from the patient's mobile device.
[0032] The processing steps of the time alignment engine are as follows: The system automatically labels the actual collection time of each data point, accurate to the minute. If the data for a key indicator (such as joint mobility to be assessed 6 weeks post-surgery) exceeds the preset time window (e.g., 72 hours before the assessment), the system filters out expired data that exceeds the preset assessment time window, triggers a data expiration warning, and then re-collects the data. For multiple time point records of the same indicator, the system automatically selects the valid data with the timestamp closest to the current assessment time.
[0033] The processed data is then entered into the corresponding fields in section O (e.g., "Active knee flexion angle: 110° (collection time: 202X-XX-XX 09:30)"). For key assessment indicators that are mandatory for this stage but missing (e.g., "dislocation risk assessment" not recorded after total hip replacement), the system provides a red mandatory reminder, enabling automatic filling and missing data alerts.
[0034] The system automatically retrieves the current patient's training completion data from the rehabilitation prescription execution database, including: the number of times / sets of each movement completed, the training completion rate (actual completion / planned completion), and whether any abnormal reactions occurred (increased pain, increased swelling, etc.). This data is automatically populated into the P (treatment plan) section (as a reference for plan execution) and A (assessment and analysis) section (as a basis for analyzing training adherence) to complete the automatic pre-population of data.
[0035] S3 retrieves objective data from the surgical procedure-stage quantitative milestone target database and automatically generates a structured evaluation and analysis draft.
[0036] After pre-filling, intelligent assisted data entry will be performed to complete the data for subjective information (S) and assessment analysis (A). In this embodiment, the system provides an efficient data entry interface. For the S part, it supports pre-filling by patients on their mobile devices (filling in chief complaint, pain VAS, sleep, and mood via WeChat mini-program before the consultation), and the data is automatically synchronized to the therapist's end. The therapist only needs to confirm or supplement key information, and it supports voice-to-text and quick input of commonly used phrases.
[0037] For Part A, the system automatically generates a structured evaluation and analysis draft based on objective data and a built-in milestone target library. The evaluation and analysis processing steps are as follows: The system retrieves quantitative milestone targets specific to each surgical procedure and stage, constructed from AOPT, AAOS guidelines, and expert consensus. It automatically compares objective data with corresponding stage target values, generating results indicating whether each indicator has met or not, and providing suggestions for improving rehabilitation training for indicators that have not met the targets. For example: "Compared to the mid-term post-ACL reconstruction target: ① Active knee flexion angle 110° (target ≥120°), not met; ② Knee extension lag 2° (target ≤3°), met; ③ Quadriceps strength 4 / 5 (target ≥4 / 5), met. Overall analysis: Joint mobility has progressed well but has not yet reached the stage target; emphasis should be placed on strengthening the end-flexion angle training." Therapists can then modify and confirm these recommendations.
[0038] S4 integrates all the data generated above in the standard SOAP format, automatically performs triple logical checks, automatically marks abnormal items and forces corrections, generates the final SOAP document, and synchronizes it to the HIS system for archiving after confirmation.
[0039] In this embodiment, the triple logic check includes integrity, reasonableness of numerical range, and cross-part logic consistency check.
[0040] Completeness verification checks whether all required fields have been filled in (such as the core assessment items in Part O, the analysis conclusions in Part A, and the plan content in Part P). Numerical range reasonableness verification checks whether the entered data is within the physiological / clinical reasonable range (such as joint range of motion 0-150°, VAS score 0-10).
[0041] Cross-part logical consistency verification is as follows: Consistency check between subjective data (S) pain VAS score and objective data (O) analgesic use records. If S records "severe pain" but O has no pain score or analgesic use record, then prompt for supplementation.
[0042] The assessment analysis (A) verifies the consistency between the conclusion of muscle strength improvement and the trend of muscle strength grade change in subjective data / objective data (S / O). If the conclusion of A is "muscle strength improved", but O shows a decrease in muscle strength grade, it indicates a contradiction.
[0043] The consistency of the treatment plan (P) action setting and assessment analysis (A) infection risk conclusion is checked. For example, if A indicates "high risk of infection", but P does not stop invasive treatment, then the risk is indicated.
[0044] The consistency of the training parameters of the treatment plan (P) with the training movements, intensity, and contraindicated movements of the current effective rehabilitation prescription is checked. The plan in the P part is automatically compared with the current effective prescription in the prescription system. If there is a discrepancy, a warning will pop up and confirmation will be required.
[0045] In this embodiment, the consistency verification between the treatment plan (P) and the current effective rehabilitation prescription is as follows: extract the training action name, number of repetitions per set, number of sets, intensity level, and list of contraindicated actions from the treatment plan, and compare them field by field with the corresponding fields of the currently effective rehabilitation prescription in the prescription system. If the fields are inconsistent, a red mandatory warning is automatically triggered. The therapist passes the verification after confirming the reasonableness of the difference.
[0046] After generating the final SOAP document, the therapist needs to confirm it electronically online, completing the document generation and signing. It can also be exported as PDF / Word with one click, printed directly for archiving, or synchronized to the hospital's HIS / EMR system.
[0047] Then, the achievement status of milestone indicators in the assessment analysis (A) (such as "knee flexion angle not met"), the analysis of the reasons for the non-met results, and the rehabilitation plan adjustment instructions in part P (such as "it is recommended to strengthen knee flexion training") are pushed to the rehabilitation prescription system in structured JSON format, which automatically triggers the switching of rehabilitation stages or the optimization of prescription content, forming a closed loop of the rehabilitation process.
[0048] All structured SOAP data is automatically incorporated into the disease-specific database and can be directly exported to SPSS / R format. It supports multi-center research analysis and quality indicator statistics (such as milestone achievement rates and complication rates for each surgical procedure and stage), enabling the management of research data in the database.
[0049] In this embodiment, by breaking down data silos such as HIS, LIS, PACS, rehabilitation assessment, wearable devices, and prescription execution, the system achieves fully automated retrieval and backfilling of objective data, completely eliminating the risk of errors and omissions caused by manual copying and pasting, achieving zero duplicate data entry, and preventing transcription errors.
[0050] Simultaneously, it adaptively matches structured SOAP templates according to surgical procedure and stage, and enforces triple checks on completeness, numerical values, and logic to ensure uniform document format, complete content, and logical consistency. This easily meets the requirements of tertiary hospital accreditation, medical insurance audits, and quality control, ensuring 100% compliance with regulations. Furthermore, it automatically compares Part A with the surgical procedure-stage quantitative milestone target database to generate structured and quantifiable assessment analyses. This completely eliminates differences in therapists' personal experience, making rehabilitation assessments comparable and traceable, and ensuring objectivity and structure in the assessment analysis.
[0051] Furthermore, this solution pioneers a dynamic consistency check between the P section and the prescription system, automatically detecting and warning of discrepancies between the "plan" and the "prescription." Simultaneously, cross-section logic checks can uncover hidden conflicts between S / O / A / P, effectively reducing potential medical errors and disputes, and lowering medical risks.
[0052] Finally, structured SOAP data is automatically imported into the database and can be exported to a research format with one click. It can easily generate departmental rehabilitation quality indicator reports (milestone achievement rate, plan adjustment distribution, complication rate) and provide closed-loop feedback to the prescription system, driving continuous optimization of rehabilitation programs and empowering continuous improvement in research and quality.
[0053] This solution significantly reduces the writing time of a complete SOAP document from 30 minutes in traditional manual work to less than 5 minutes, increasing efficiency by more than 6 times. It frees therapists from tedious paperwork, allowing them to focus on patient care and greatly improving document efficiency.
[0054] The above descriptions are merely embodiments of the present invention, and common knowledge such as specific technical solutions and / or characteristics are not described in detail here. It should be noted that those skilled in the art can make various modifications and improvements without departing from the technical solutions of the present invention, and these should also be considered within the scope of protection of the present invention. These modifications and improvements will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.
Claims
1. A SOAP document generation method based on surgical procedure staging and multi-source data linkage, characterized in that, Includes the following steps: S1: Obtain patient identity information and rehabilitation scenario, and dynamically determine the patient's surgical procedure and current rehabilitation stage by combining multi-dimensional data; Match and retrieve the corresponding SOAP template from the pre-set structured document template library. The SOAP template has four types of specialized feature chapters. S2 retrieves multi-source heterogeneous data, filters and evaluates the latest valid data within the time window using the time alignment engine, and automatically fills it into the objective data section; it also retrieves the current patient's execution completion data and fills it into the treatment plan and evaluation analysis sections. S3 retrieves objective data from the surgical procedure-stage quantitative milestone target database and automatically generates a structured evaluation and analysis draft; S4 integrates the data generated above in SOAP format, performs triple logical checks, automatically marks abnormal items and forces corrections, generates the final SOAP document, and synchronizes it to the HIS system for archiving after confirmation.
2. The SOAP document generation method based on surgical procedure staging and multi-source data linkage according to claim 1, characterized in that, The dynamic determination steps for rehabilitation staging are as follows: The system calls upon a pre-defined staging rule library specific to each surgical procedure, taking the patient's postoperative days, wound healing level, active joint range of motion measurement, manual muscle strength assessment, and weight-bearing capacity as input parameters. Through rule matching, it automatically determines the current rehabilitation stage, which includes the healing and repair stage, early rehabilitation stage, strength and control stage, and return to exercise stage.
3. The SOAP document generation method based on surgical procedure staging and multi-source data linkage according to claim 1, characterized in that, The multi-source heterogeneous data includes: surgical records, vital signs, and comorbidity information from the HIS system; CRP, blood routine, and inflammatory factor test indicators from the LIS system; postoperative joint alignment angles and implant position image measurements from the PACS system; joint range of motion, muscle strength grading, balance scores, and 6-minute walking distance data from rehabilitation assessment equipment; step count, activity intensity, and sleep duration data from wearable devices; number of training movements completed, number of sets, completion rate, and abnormal reaction records from the prescription execution system; and home self-test data on the patient's active joint range of motion and single-leg standing time from mobile devices.
4. The SOAP document generation method based on surgical procedure staging and multi-source data linkage according to claim 1, characterized in that, The four categories of specialized features include: Subjective data section: defined as S, pre-defined inquiry points related to the procedure type and stage; Objective data section: defined as O, a list of objective indicators that must be evaluated in each phase; Evaluation and Analysis Section: Defined as A, embedding the quantitative milestone target library of the surgical procedure-staging; Treatment plan section: Defined as P, with reserved data fields for linkage with the prescription system.
5. The SOAP document generation method based on surgical procedure staging and multi-source data linkage according to claim 1, characterized in that, In S2, the processing steps of the time alignment engine are as follows: Automatically labels the actual collection time of each data point, accurate to the minute; filters out expired data whose collection time exceeds the preset evaluation time window, and triggers a data expiration warning before re-collecting; for multiple time point records of the same indicator, automatically selects the valid data with the timestamp closest to the current evaluation time.
6. The SOAP document generation method based on surgical procedure staging and multi-source data linkage according to claim 1, characterized in that, In S3, the evaluation and analysis processing steps are as follows: The system retrieves a quantitative milestone target library specific to each surgical procedure and stage, constructed from AOPT, AAOS guidelines, and expert consensus. It automatically compares objective data with the corresponding stage target values, generates a result indicating whether each indicator meets or fails to meet the target, and provides suggestions for improving rehabilitation training for indicators that do not meet the target.
7. The SOAP document generation method based on surgical procedure staging and multi-source data linkage according to claim 1, characterized in that, In S4, the triple logic check includes integrity, numerical range reasonableness, and cross-part logic consistency check; the cross-part logic consistency check is as follows: The consistency of subjective data (VAS pain score) with objective data (analgesic use records), the consistency of assessment and analysis of muscle strength improvement conclusions with objective data (muscle strength grading change trends), the consistency of treatment plan movement settings with assessment and analysis of infection risk conclusions, and the consistency of treatment plan training parameters with current effective rehabilitation prescription training movements, intensity, and contraindications.
8. The SOAP document generation method based on surgical procedure staging and multi-source data linkage according to claim 7, characterized in that, The consistency verification between the treatment plan and the current effective rehabilitation prescription is as follows: extract the training action name, number of repetitions per set, number of sets, intensity level, and list of contraindicated actions from the treatment plan, and compare them field by field with the corresponding fields of the currently effective rehabilitation prescription in the prescription system. If the fields are inconsistent, a red mandatory warning is automatically triggered. The therapist confirms the reasonableness of the difference and passes the verification.
9. The SOAP document generation method based on surgical procedure staging and multi-source data linkage according to claim 1, characterized in that, In S4, after generating the final SOAP document, the achievement status of milestone indicators in the assessment analysis, the analysis of reasons for non-achievement indicators, and the instructions for adjusting the rehabilitation plan are pushed to the rehabilitation prescription system in a structured JSON format, which automatically triggers the switching of rehabilitation stages or the optimization of prescription content, forming a closed loop in the rehabilitation process.
10. The SOAP document generation method based on surgical procedure staging and multi-source data linkage according to claim 7, characterized in that: The integrity check is to check whether all required fields have been filled in; the numerical range reasonableness check is to check whether the entered data is within the physiological / clinical reasonable range.