A HIS-based intelligent diagnosis and decision support method for coronary heart disease angina pectoris TCM syndrome elements

Abstract

The application discloses a kind of intelligent diagnosis and decision support method of coronary heart disease angina TCM syndrome factor based on HIS, it is related to medical information and TCM diagnosis and treatment technical field, the method will coronary heart disease angina TCM syndrome factor diagnostic criteria be deconstructed into standardization diagnostic item, build machine-readable weight rule base containing score, threshold and severity grading rule, response syndrome differentiation operation generates electronic collection form, receives structured data storage to the TCM syndrome factor fact table associated with patient main index and medical record, after structured data is submitted, trigger intelligent computing engine verification essential item, total score is accumulated, determine the establishment of syndrome factor and severity, through TCM diagnosis and treatment knowledge graph push personalized treatment or nursing suggestion in clinical business node, and continuously update rule base and knowledge graph through clinical data feedback.The application integrates corresponding function module to realize and deeply fuse with hospital information system, improves the standardization and efficiency of diagnosis and treatment, and has strong practicality.

Description

Technical Field

[0001] This invention relates to the fields of medical information technology and traditional Chinese medicine diagnosis and treatment, specifically to a method for intelligent diagnosis and decision support of TCM syndrome elements in coronary heart disease and angina pectoris based on HIS. Background Technology

[0002] Traditional Chinese medicine (TCM) possesses a unique theoretical system and rich clinical experience in the diagnosis and treatment of coronary heart disease and angina pectoris. Its core lies in the identification of "syndrome elements" (such as blood stasis, qi deficiency, and phlegm turbidity). Digitizing and intelligentizing the TCM syndrome element identification process and integrating it with modern hospital information systems (HIS) is of great significance for improving the standardization, reproducibility, and clinical research level of TCM diagnosis and treatment. In recent years, research has explored the application of artificial intelligence technology to the auxiliary diagnosis of TCM syndrome elements, such as using quantitative methods to analyze characteristics like tongue coating and pulse, or constructing predictive models from symptoms to syndrome elements. These efforts lay the foundation for the objectification and standardization of TCM.

[0003] In existing technical solutions, related patents mainly develop in two directions: one is to develop independent TCM intelligent diagnostic models or systems, and the other is to construct TCM and Western medicine auxiliary diagnostic tools for coronary heart disease. For example, a method, system, and storage medium for intelligent identification of TCM syndrome elements (publication number CN119252470B) constructs a decision tree model by analyzing the frequency and correlation coefficient of symptom manifestation categories to achieve intelligent identification of syndrome elements. Another prior art, a patent for a method, device, and electronic device for generating auxiliary diagnostic information for coronary heart disease (publication number CN117334330A), proposes to generate auxiliary diagnostic information by extracting and fusing feature vectors from multiple sources of descriptive information such as the patient's chief complaint, tongue appearance, and pulse. These technical solutions achieve auxiliary diagnostic functions within a specific scope.

[0004] However, a deeper analysis of existing technologies reveals a significant limitation: most of these solutions exist as supplementary tools or isolated models independent of the hospital's core business processes. The structured diagnostic information or data models they generate are not deeply and natively integrated with the hospital information system that supports the entire patient care cycle at the data, business logic, and application layers. This prevents the key data generated during TCM diagnosis from being transformed into core data units that the hospital information system can identify, calculate, and participate in business processes. Consequently, this disrupts the decision-making process within the modern medical closed loop constructed by the hospital information system, from diagnosis to treatment and nursing. The system struggles to automatically trigger and push personalized treatment recommendations at key clinical business nodes such as clinical diagnosis and prescription issuance based on real-time and structured syndrome element diagnostic results. This means that TCM treatment decisions remain highly dependent on physicians' offline personal experience, failing to achieve a complete digital closed loop from data collection to intelligent analysis and deep integration with specific clinical scenarios for decision support. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and provide an intelligent diagnosis and decision support method for TCM syndrome elements of coronary heart disease and angina pectoris based on HIS. By standardizing and modeling TCM syndrome elements of coronary heart disease and angina pectoris and constructing a weighted rule base, combined with dynamic form collection of structured data and intelligent computing engine for automated diagnosis, and linking TCM diagnosis and treatment knowledge graph with the core business processes of hospital information system, it can achieve deep native integration of TCM syndrome data and HIS, forming a complete digital closed loop from data collection, intelligent diagnosis to personalized treatment plan push, and solving the problems of disconnection between existing technologies and hospital business processes and interruption of decision-making processes.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: On one hand, a method for intelligent diagnosis and decision support of TCM syndrome elements in coronary heart disease and angina pectoris based on HIS, the method comprising the following steps:

[0007] Step 1, Digital Modeling and Dynamic Rule Base Construction of Syndrome Elements: Based on the diagnostic criteria for syndrome elements of coronary heart disease and angina pectoris, the definition of the target TCM syndrome elements is deconstructed into multiple diagnostic items. A globally unique system code is assigned to each diagnostic item and a standardized data format is defined. Based on the quantitative scoring system of the diagnostic criteria, a weighted rule base stored in a machine-readable format is constructed. The weighted rule base defines the score, essential diagnostic items, diagnostic threshold, and severity grading rules for each diagnostic item.

[0008] Step 2, Dynamic Structured Data Acquisition and Persistent Storage: In response to the diagnostic operation initiated by the doctor's workstation in the hospital information system, the weight rule library is called to dynamically render and generate the corresponding electronic data acquisition form interface, receive the structured data input through the form, and persistently store the structured data in the newly added TCM syndrome element fact table associated with the patient's master index and medical records in the hospital information system database.

[0009] Step 3, Automated diagnosis by the intelligent computing engine: After the structured data is submitted, the intelligent computing engine service for syndrome elements deployed in the background is automatically triggered. The intelligent computing engine service for syndrome elements loads the corresponding weight rule library, verifies whether the essential diagnostic conditions are met, maps and accumulates the total score according to the weight rule library, compares the total score with the diagnostic threshold, automatically determines whether the target TCM syndrome element is established and its severity level, generates a structured diagnostic report and writes it back to the TCM syndrome element fact table;

[0010] Step 4, Knowledge Graph-Driven Intelligent Decision Support: At the preset clinical business nodes of the hospital information system, the latest syndrome element diagnosis results of patients are actively obtained. Using the syndrome element diagnosis results as input, a pre-constructed TCM diagnosis and treatment knowledge graph is queried. The TCM diagnosis and treatment knowledge graph stores the semantic relationships between syndrome elements, treatment methods, prescriptions, and Chinese herbal entities. In the interface of the clinical business node, personalized treatment plans or nursing suggestions generated based on the reasoning of the TCM diagnosis and treatment knowledge graph are pushed.

[0011] Furthermore, in step one, the diagnostic items include core symptom items, tongue appearance items, pulse appearance items, objective signs items, and quantitative indicator items;

[0012] The core symptom item is defined as a necessary item for diagnosis, and the data corresponding to the core symptom item is a Boolean value.

[0013] The data for the tongue and pulse entries are classification values ​​for predefined options;

[0014] The data for the quantitative indicator entries are level values.

[0015] Furthermore, in step three, the automated diagnostic steps of the intelligent computing engine specifically include:

[0016] The intelligent calculation engine service for syndrome elements first checks whether the data of the corresponding diagnostic essential items in the submitted structured data meet the preset conditions. If not, the process is terminated.

[0017] If the condition is met, all submitted diagnostic entries are traversed, and scores are mapped and accumulated according to the scoring rules defined in the weight rule library to obtain the total score.

[0018] The total score is compared with the diagnostic threshold defined in the weighted rule base to determine whether the syndrome elements are valid;

[0019] The severity level of the syndrome element is determined based on whether the total score falls within the severity grading interval defined in the weighted rule base.

[0020] Furthermore, in step four, the preset clinical business node includes a medical order issuance interface;

[0021] On the medical order issuance interface, the diagnostic results of the syndrome elements are obtained through event listening. Based on the correlation between syndrome elements and treatment methods, and treatment methods and prescriptions in the TCM diagnosis and treatment knowledge graph, a personalized treatment plan that includes at least recommended prescriptions or Chinese patent medicines is generated and displayed on the side of the medical order issuance interface in the form of a prompt box or recommendation list.

[0022] Furthermore, the method also includes:

[0023] Step 5, Clinical Data Feedback and Knowledge Evolution: Through the clinical data insight and analysis module, the structured diagnostic data accumulated in the TCM syndrome element fact table is analyzed regularly to explore the correlation patterns between symptom combinations and syndrome element diagnostic results, and knowledge update suggestions are generated. Through the knowledge rule collaborative management platform, the knowledge update suggestions are received and reviewed by experts. The approved new knowledge is then synchronously updated to the TCM diagnosis and treatment knowledge graph and the weight rule base.

[0024] On the other hand, a HIS-based intelligent diagnosis and decision support system for TCM syndrome elements in coronary heart disease and angina pectoris is applicable to a HIS-based intelligent diagnosis and decision support method for TCM syndrome elements in coronary heart disease and angina pectoris. This system integrates the following modules into the hospital information system:

[0025] The syndrome element standard knowledge base module is used to store and manage a digital weighted rule base for all TCM syndrome elements;

[0026] The dynamic form service engine module is used to dynamically generate a standardized data collection form interface on the front end based on the weight rule base definition in the syndrome element standard knowledge base module.

[0027] The intelligent calculation engine module for syndrome elements is deployed in the form of microservices. It is used to load the weight rule library, perform automated syndrome differentiation calculations, and output structured diagnostic reports.

[0028] The TCM diagnosis and treatment knowledge graph module is used to store the TCM clinical knowledge network constructed by ontology and supports semantic query and reasoning based on the diagnostic results of syndrome elements.

[0029] The business scenario intelligent integration interface module is used to monitor business events in the hospital information system and push the diagnostic results output by the syndrome element intelligent calculation engine module and the recommendation information generated by the traditional Chinese medicine diagnosis and treatment knowledge graph module to the business terminal interface of the medical order system and the nursing system.

[0030] Furthermore, the weight rule base stored in the syndrome element standard knowledge base module adopts JSON or XML format. The weight rule base includes syndrome element codes, syndrome element names, definitions of essential diagnostic items, scoring rules for each diagnostic item, diagnostic thresholds, and severity grading rules.

[0031] Furthermore, the system also includes a clinical data insight and analysis module, which serves as a background service to periodically perform association rule analysis on the structured diagnostic data in the TCM syndrome element fact table to uncover high-confidence symptom-syndrome element association patterns and generate structured knowledge update suggestion sheets.

[0032] Furthermore, the system also includes a knowledge rule collaborative management platform, which provides a visual management interface for receiving knowledge update suggestion forms generated by the clinical data insight and analysis module, organizing expert reviews, and converting the new knowledge confirmed by the experts into new relationships or entities in the TCM diagnosis and treatment knowledge graph module, or synchronously updating the corresponding weight rule base configuration in the syndrome element standard knowledge base module.

[0033] Furthermore, the system also includes a clinical pathway quality control rule library for syndrome elements, which stores quality control rules defined in a machine-readable format. The business scenario intelligent integration interface module is also used to compare the patient's real-time syndrome element diagnosis results with the rules in the clinical pathway quality control rule library for syndrome elements in the clinical pathway management process of the hospital information system, and to trigger mandatory prompts or warning information on the business terminal interface when deviations from the rules are detected.

[0034] Compared with existing technologies, this HIS-based intelligent diagnosis and decision support method for TCM syndrome elements in coronary heart disease and angina pectoris has the following beneficial effects:

[0035] I. This invention deconstructs the diagnostic criteria for TCM syndrome elements in coronary heart disease and angina pectoris into standardized diagnostic items, constructs a machine-readable weighted rule base, and achieves deep native integration with hospital information systems at the data layer, business logic layer, and application layer. Relying on dynamic forms to collect structured data, the intelligent computing engine automatically completes the diagnosis of syndrome elements. Personalized treatment plans are proactively pushed at key clinical nodes such as clinical diagnosis and prescription, forming a complete digital closed loop from data collection, intelligent analysis to decision support. This effectively solves the problems of disconnection between TCM diagnosis tools and core hospital business processes and interruption of decision-making processes in existing technologies, making TCM diagnosis data a core data unit that can participate in business flow in hospital information systems, and improving the standardization and repeatability of TCM diagnosis and treatment.

[0036] Second, this invention establishes a clinical data feedback and knowledge evolution mechanism to periodically mine the correlation patterns between symptom combinations and syndrome elements in diagnostic results. After expert review, the weight rule base and TCM diagnosis and treatment knowledge graph are updated simultaneously to achieve continuous optimization of diagnosis and treatment knowledge. At the same time, it integrates a clinical pathway quality control rule base to conduct real-time comparison and early warning of the diagnosis and treatment process, standardize the TCM syndrome differentiation process, and improve the accuracy of syndrome differentiation and clinical efficacy.

[0037] Other advantages, objectives and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination or study, or may be learned from the practice of the invention. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0039] Figure 1 This is a schematic diagram of the overall system architecture and data flow of the present invention;

[0040] Figure 2 This is a flowchart of the intelligent diagnosis process for syndrome elements of the present invention;

[0041] Figure 3 This is the knowledge graph-driven decision-making and evolution closed-loop diagram of the present invention. Detailed Implementation

[0042] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0043] Example 1

[0044] like Figures 1 to 3 As shown, this embodiment uses common blood stasis syndrome elements in coronary heart disease and angina pectoris as the application object, and elaborates on the implementation process of the technical solution of the present invention in conjunction with the actual clinical scenario of a hospital information system. The core of this embodiment is to realize intelligent diagnosis and decision support of TCM syndrome elements through a three-layer architecture of data model standardization, business logic engineization, and decision application scenarioization. The following is a detailed description in conjunction with specific implementation methods.

[0045] In this embodiment, the implementation prerequisites and environment configuration are as follows:

[0046] The operating environment of this embodiment is based on a mainstream hospital information system. This system has core functions such as patient master index management, medical record storage, medical order issuance, and nursing plan formulation. The database adopts a relational database management system, which supports structured data storage and related queries. The front end supports dynamic form rendering and event listening, and the back end supports microservice deployment and interface calls.

[0047] At the system integration level, the new module of this invention achieves deep integration with the core database, business logic layer and application layer of the hospital information system through standard interfaces. It does not change the core architecture and business processes of the original system, but only adds a TCM syndrome element fact table at the data layer, deploys a syndrome element intelligent calculation engine at the logic layer, and expands the dynamic collection form and decision support interface at the application layer.

[0048] In this embodiment, the digital modeling of syndrome elements and the construction of a dynamic rule base are as follows:

[0049] Structured breakdown of diagnostic items:

[0050] Based on the diagnostic criteria for angina pectoris syndrome elements in coronary heart disease, the clinical diagnostic basis of blood stasis syndrome elements is broken down into five categories of diagnostic items: core symptoms, tongue appearance, pulse appearance, objective signs, and quantitative indicators. Each item is assigned a globally unique system code, and the data format and collection method are clearly defined to ensure the standardization and computability of the data.

[0051] The core symptom item is essential for diagnosis, defining the core diagnostic boundaries of blood stasis syndrome elements. The data format is Boolean, indicating the presence of the symptom. Tongue and pulse items use predefined classification values ​​to eliminate ambiguity in clinical descriptions and ensure data consistency. Objective sign items use Boolean format, breaking down complex signs into independent atomized items for accurate data collection. Quantitative indicator items use ordinal values ​​to achieve semi-quantitative assessment and improve diagnostic objectivity. Specific diagnostic item settings are as follows:

[0052] The primary symptom is fixed chest pain, the system code is SXXY001, the data format is Boolean, and the data acquisition method is selected by checking the box.

[0053] Tongue appearance: The tongue is dark purple or has petechiae and ecchymosis. System code: TSXY001. Data format: classification value. Collection method: drop-down selection. Selectable values: dark purple, petechiae, ecchymosis, normal.

[0054] Tongue appearance: Dark purple veins under the tongue, system code TSXY002, data format level value, acquisition method slider selection, selectable values ​​0, 1, 2, 3;

[0055] Pulse characteristics: Choppy pulse, system code MXXY001, data format classification value, acquisition method drop-down selection, selectable values: choppy, intermittent, normal;

[0056] Objective signs: Dark purple complexion, system code SXXY0031, data format Boolean, collection method selected;

[0057] Objective signs: petechiae or ecchymosis on the body, system code SXXY0032, data format Boolean, collection method selected;

[0058] Objective signs: numbness in limbs, system code SXXY0033, data format Boolean, collection method selected;

[0059] Objective signs: Dark purple or dark red lips, system code SXXY0034, data format Boolean, collection method selected.

[0060] Construction of the weighted rule library:

[0061] The weighted rule library is stored in JSON format and includes basic information on syndrome elements, definitions of essential diagnostic items, scoring rules, diagnostic thresholds, and severity grading rules, ensuring machine readability and execution. The weighted rules are set based on the quantitative scoring system of the diagnostic criteria for angina pectoris syndrome elements in coronary heart disease, combined with clinical expert consensus to determine the weight of each item. Diagnostic thresholds and severity grading ranges were determined through a retrospective analysis of clinical data from 1000 patients with angina pectoris in coronary heart disease, ensuring the clinical applicability of the rules.

[0062] The specific contents of the weighted rule base are as follows:

[0063] Syndrome element code XY01, syndrome element name blood stasis syndrome, standard source: diagnostic criteria for syndrome elements of coronary heart disease angina pectoris;

[0064] Essential diagnostic criteria: Code SXXY001, description of fixed chest pain must be present;

[0065] Diagnostic threshold 7;

[0066] Severity grading: mild (minimum score 7, maximum score 11), moderate (minimum score 12, maximum score 16), severe (minimum score 17, maximum score 21);

[0067] Scoring rules:

[0068] Code SXXY001, name chest fixed pain, score 4, rule: present counts 4 points, absent counts 0 points;

[0069] Code TSXY001, name purple or dull tongue or petechiae and ecchymosis on the tongue body, score mapping: purple dull 4, petechiae 4, ecchymosis 4, normal 0;

[0070] Code TSXY002, name purple dull sublingual vein, score mapping: 0→0, 1→1, 2→2, 3→3;

[0071] Code SXXY0031, name purple dull complexion, score 3, rule: present counts 3 points, absent counts 0 points;

[0072] Code SXXY0032, name petechiae or ecchymosis on the body, score 3, rule: present counts 3 points, absent counts 0 points;

[0073] Code SXXY0033, name limb numbness, score 2, rule: present counts 2 points, absent counts 0 points;

[0074] Code SXXY0034, name purple dull or dark red lips, score 2, rule: present counts 2 points, absent counts 0 points;

[0075] Code MXXY001, name unsmooth pulse, score mapping: unsmooth 2, intermittent pulse 2, normal 0.

[0076] In this embodiment, the dynamic structured data collection and persistent storage are as follows:

[0077] Dynamic form generation:

[0078] In the electronic medical record interface of the doctor workstation in the hospital information system, a traditional Chinese medicine syndrome differentiation function entry is set. When a doctor clicks on the blood stasis syndrome assessment option, the system triggers the dynamic form service engine module. This module automatically renders and generates a standardized collection form interface by reading the diagnostic entry information in the weight rule library. The form controls correspond one-to-one with the data formats of the diagnostic entries: boolean type entries use checkbox controls, categorical value entries use dropdown selection controls, and graded numerical entries use slider controls.

[0079] The layout of the form interface follows the clinical diagnosis and treatment habits, and is arranged in the order of core main symptoms, tongue manifestation, pulse condition, objective signs, and quantitative indicators. The corresponding item names are marked beside each control for the doctor to quickly identify and enter. The form supports real-time verification function. When the doctor does not check the core main symptom item, the system will prompt to confirm the core main symptom first to avoid entering invalid data.

[0080] Data collection and submission:

[0081] Based on the actual clinical manifestations of the patient, the doctor completes data entry through the form controls. Taking a certain patient as an example, whose clinical symptoms are the presence of fixed chest pain, petechiae on the tongue, grade 2 of dark purple sublingual veins, the presence of dark purple complexion, no petechiae or ecchymosis on the body, the presence of limb numbness, the presence of dark purple lips, and a stringy pulse, the doctor's corresponding entry operations are as follows:

[0082] Check the control for the core main symptom of fixed chest pain;

[0083] Select "petechiae" from the drop-down box of tongue manifestation of purple or dark tongue or petechiae or ecchymosis on the tongue body;

[0084] Drag the slider of dark purple sublingual veins to position 2;

[0085] Check the control for the objective sign of dark purple complexion;

[0086] Do not check the control for the objective sign of petechiae or ecchymosis on the body;

[0087] Check the control for the objective sign of limb numbness;

[0088] Check the control for the objective sign of dark purple or dull red lips;

[0089] Select "stringy" from the drop-down box of pulse condition of stringy pulse.

[0090] After the entry is completed, the doctor clicks the submit button, and the system encapsulates the entered data into a structured data object. The data object includes the patient's main index, visit record index, syndrome element code, and the entered values of each diagnosis item to ensure the integrity and relevance of the data.

[0091] Data persistent storage:

[0092] A fact table of traditional Chinese medicine syndrome elements is newly added to the hospital information system database. This table is associated with the core main tables such as the patient main index table, visit record table, and doctor's advice table through key fields to achieve cross-table query and fusion analysis of data. The field design of the fact table of traditional Chinese medicine syndrome elements is as follows:

[0093] Record number: uniquely identifies each syndrome element diagnosis record, data type is string;

[0094] Patient main index: the key field associated with the patient main index table, data type is string;

[0095] Medical record index: Relates to the key fields of the medical record table; data type is string.

[0096] Syndrome element encoding: Identifies the corresponding syndrome element; data type string.

[0097] Syndrome element name: Data type string;

[0098] Diagnostic entry data: Stores the entered values ​​for each diagnostic entry; the data type is structured data.

[0099] Diagnostic results: Stores the results of automated diagnosis; the data type is structured data.

[0100] Record time: The time when data entry and diagnosis were completed, and the data type timestamp.

[0101] The system writes the encapsulated structured data objects into the TCM syndrome element fact table and simultaneously creates a data index to improve data query efficiency. The data storage process employs a transaction mechanism to ensure the atomicity of data writes and prevent data loss or inconsistency.

[0102] In this embodiment, the automated diagnosis of the intelligent computing engine is as follows:

[0103] Engine triggering and data loading:

[0104] Once structured data is submitted and stored in the TCM syndrome element fact table, the system automatically triggers the syndrome element intelligent calculation engine service through database triggers. This engine is deployed as a microservice, supports high-concurrency processing, and can quickly respond to diagnostic requests initiated by multiple physicians simultaneously.

[0105] The engine first loads the corresponding weight rule library from the syndrome element standard knowledge base module based on the syndrome element encoding in the structured data, and at the same time reads the patient's diagnostic entry data from the TCM syndrome element fact table to prepare for diagnostic calculation.

[0106] Diagnostic algorithm and calculation process:

[0107] The intelligent computing engine uses a weighted summation algorithm to perform diagnostic calculations for syndrome elements. The algorithm process consists of four steps: verification of essential items, score accumulation, threshold comparison, and severity classification, ensuring the objectivity and repeatability of the diagnostic process.

[0108] Required verification:

[0109] The engine first checks the value of the core symptom entry in the diagnostic entry data. If the core symptom does not exist, it directly determines that the blood stasis syndrome element is not valid, terminates the diagnostic process, generates a diagnosis result of invalidity, and writes it back to the TCM syndrome element fact table; if the core symptom exists, it proceeds to the score accumulation step.

[0110] Score accumulation algorithm:

[0111] The scores are accumulated using a weighted summation algorithm, and the calculation formula is as follows:

[0112]

[0113] in: The total score for the diagnostic items. In this embodiment, the total number of diagnostic entries is [number]. , For the first The weight score for each diagnostic item is defined by the weight rule base. For the first The coefficients for each diagnostic entry, when Boolean entries exist. When it does not exist Category value entries and grade value entries Set the score mapping coefficient corresponding to the entered value.

[0114] Based on the patient data in this embodiment, the score calculation process is as follows:

[0115] The core symptom is fixed chest pain: , ,Score ;

[0116] Tongue appearance: The tongue body is purplish-dark or has petechiae or ecchymoses. , ,Score ;

[0117] Dark purple veins under the tongue: This corresponds to level 2, with a score of 2. ,Score ;

[0118] Objective physical signs: purplish-dark complexion , ,Score ;

[0119] Objective signs include petechiae or ecchymosis on the body: , ,Score ;

[0120] Objective signs: numbness in the limbs: , ,Score ;

[0121] Objective signs: Dark purple or dark red lips , ,Score ;

[0122] Pulse texture: (The pulse is rough and hesitant.) , ,Score ;

[0123] Total Score .

[0124] Threshold comparison and severity grading:

[0125] The engine compares the calculated total score with the diagnostic threshold in the weighted rule base. If the total score is lower than the diagnostic threshold, the blood stasis syndrome element is determined to be invalid. If the total score is higher than or equal to the diagnostic threshold, the blood stasis syndrome element is determined to be valid, and the grading result is determined according to the severity grading interval in which the total score is located.

[0126] In this embodiment, the diagnostic threshold is 7. The patient's total score of 19 is higher than the diagnostic threshold, and the blood stasis syndrome element is determined to be valid. The severity grading range is mild (7 to 11), moderate (12 to 16), and severe (17 to 21). Since 19 falls within the range of 17 to 21, the grading result is severe.

[0127] Diagnostic result generation and write-back:

[0128] The engine generates a structured diagnostic report, which includes information such as syndrome element codes, syndrome element names, diagnostic results, total score, severity, diagnostic threshold, scoring details, and recording time. The diagnostic report is then written back to the diagnostic result field of the TCM syndrome element fact table, and simultaneously fed back in real-time to the electronic medical record interface of the hospital information system's doctor workstation for physicians to view and confirm.

[0129] In this embodiment, the knowledge graph-driven intelligent decision support is as follows:

[0130] Construction of a Knowledge Graph for Traditional Chinese Medicine Diagnosis and Treatment:

[0131] The TCM diagnostic and treatment knowledge graph uses a graph structure to store TCM clinical knowledge, including entities such as syndrome elements, treatment methods, prescriptions, Chinese herbal medicines, and acupoints, as well as semantic relationships between entities. The definition of entities is based on TCM theory and clinical guidelines, and the establishment of relationships is based on expert consensus and clinical practice data, ensuring the accuracy and practicality of the knowledge.

[0132] The entity attributes of a knowledge graph include information such as encoding, name, efficacy, indications, usage, and dosage. Relationships include the treatment method corresponding to the syndrome element, the prescription corresponding to the treatment method, and the traditional Chinese medicine contained in the prescription. Taking a knowledge graph fragment related to blood stasis syndrome elements as an example:

[0133] Syndrome element: Blood stasis syndrome, code XY01; Associated treatment principle: activating blood circulation, removing blood stasis, unblocking meridians and relieving pain, code ZF001.

[0134] Treatment principle: Invigorate blood circulation, remove blood stasis, unblock the meridians and relieve pain. Code: ZF001. Related formulas: Xuefu Zhuyu Decoction (code: FJ001), Compound Danshen Dripping Pills (code: FJ002), Tongxinluo Capsules (code: FJ003).

[0135] The formula Xuefu Zhuyu Decoction, code FJ001, contains Chinese herbs including peach kernel, safflower, angelica, rehmannia root, chuanxiong rhizome, red peony root, achyranthes root, platycodon root, bupleurum root, immature bitter orange peel, and licorice root, with a standard dosage of 9 grams of each. It is prepared by decocting in water and taken once a day in two divided doses.

[0136] The formula is Compound Danshen Dripping Pills, code FJ002. The dosage is 10 pills taken sublingually, 3 times a day.

[0137] The formula is Tongxinluo Capsules, code FJ003. The dosage is 2 to 4 capsules orally each time, 3 times a day.

[0138] Decision support triggers and inference:

[0139] When a physician enters the medical order issuance interface of the hospital information system, the business scenario intelligent integration interface module obtains the patient's latest syndrome element diagnosis result—severe blood stasis syndrome—through event listening. The system uses this diagnosis result as a query condition, calls the reasoning interface of the traditional Chinese medicine diagnosis and treatment knowledge graph, and performs semantic reasoning.

[0140] The reasoning process is based on association rules. First, it searches for the corresponding treatment method based on the syndrome elements, and then searches for the corresponding prescription based on the treatment method, forming a reasoning link between syndrome elements, treatment method, and prescription. At the same time, the system combines information such as the patient's consultation type and the severity of the condition to prioritize the recommended prescriptions. When the severity of the condition is severe, traditional Chinese medicine prescriptions are recommended first, followed by prepared Chinese medicines.

[0141] Decision support information push:

[0142] After the reasoning is completed, the system displays the personalized treatment plan in a recommended list on the side of the medical order issuance interface. The recommended list includes information such as recommended treatment methods, recommended prescriptions, and usage and dosage. The specific display content is as follows:

[0143] Intelligent prompt: The current patient's diagnosis is severe blood stasis syndrome. Recommended treatment: Activate blood circulation, remove blood stasis, unblock meridians, and relieve pain. Recommended formula 1: Modified Xuefu Zhuyu Decoction, dosage: decocted in water, one dose daily, divided into two administrations. Recommended formula 2: Compound Danshen Dripping Pills, dosage: 10 pills sublingually, three times daily. Recommended formula 3: Tongxinluo Capsules, dosage: 3 capsules orally, three times daily.

[0144] The recommended list supports a one-click fill function. When a doctor clicks the fill button after a recommended prescription, the system automatically generates a draft prescription with the prescription's composition, standard dosage, decoction method or usage and dosage. Doctors can then adjust the draft according to the patient's specific situation before confirming and submitting it, thus improving the efficiency of prescription writing.

[0145] In this embodiment, clinical data feedback and knowledge evolution are as follows:

[0146] Clinical data mining:

[0147] The Clinical Data Insights and Analysis module, serving as a backend service, periodically performs association rule analysis on the structured diagnostic data in the TCM syndrome element fact table. It employs the Apriori algorithm to mine association patterns between symptom combinations and syndrome element diagnostic results. The minimum support for the Apriori algorithm is set to 5%, and the minimum confidence level to 80%, ensuring that the mined association patterns have statistical significance and clinical value.

[0148] The algorithm flow is as follows:

[0149] Scan the diagnostic data in the TCM syndrome element fact table, count the frequency of each diagnostic item, and generate a frequent 1-itemset;

[0150] Candidate 2-itemsets are generated based on frequent 1-itemsets. The support of the candidate 2-itemsets is calculated by scanning the data, and frequent 2-itemsets are selected.

[0151] Repeat the above process to generate frequent 3-itemsets, frequent 4-itemsets, and so on, until no new frequent itemsets can be generated.

[0152] Extract association rules from the frequent itemset, calculate the confidence of the rules, and filter out association rules with a confidence higher than the minimum confidence.

[0153] The algorithm uncovered a correlation pattern, such as a sublingual vein purplish-darkness level greater than or equal to 2, with a 92% confidence level in its association with severe blood stasis syndrome. This correlation pattern serves as a knowledge update suggestion.

[0154] Knowledge updates and synchronization:

[0155] The system generates structured knowledge update suggestion forms from the discovered association patterns and pushes them to the knowledge rule collaborative management platform. Traditional Chinese medicine experts review the suggestion forms through the platform's visual interface. After confirming the clinical rationale of the association patterns, the system automatically converts them into new associations in the knowledge graph, strengthening the association weight between sublingual vein purplish discoloration and severe blood stasis syndrome.

[0156] Meanwhile, the system prompted the administrator to update the weighted rule base, adjusting the score for the sublingual vein purplish-darkness level greater than or equal to 2 to 4 points. The updated weighted rule base was activated in some departments through a gray-scale release. The system automatically compared the diagnostic consistency rate and clinical adoption rate under the new and old rules. After the diagnostic consistency rate of the new rules reached 90% and the clinical adoption rate reached 85%, it was officially activated throughout the hospital.

[0157] In this embodiment, the proactive quality control and early warning of the clinical pathway are as follows:

[0158] Construction of the quality control rule base:

[0159] The clinical pathway quality control rule library for syndrome elements stores quality control rules in the Drools rule engine format. The rules are formulated based on the integrated traditional Chinese and Western medicine clinical pathway and nursing safety guidelines for angina pectoris in coronary heart disease, ensuring the compliance and operability of the rules. Taking the quality control rules related to blood stasis syndrome as an example:

[0160] Rule 1: If the patient is diagnosed with moderate or severe blood stasis, a prescription for promoting blood circulation and removing blood stasis must be issued within 24 hours.

[0161] Rule 2: If the patient is diagnosed with blood stasis syndrome, the sublingual vein purplish-darkness level must be assessed daily during hospitalization;

[0162] Rule 3: If a patient's Qi deficiency score rises by more than 5 points within 8 hours, an alert for increased risk of heart failure and falls will be triggered.

[0163] Quality control comparison and early warning triggering:

[0164] Once a patient enters the clinical pathway for coronary heart disease and angina, the intelligent integration interface module for business scenarios obtains the diagnostic results of their syndrome elements in real time and compares them with the rules in the clinical pathway quality control rule library for syndrome elements in real time.

[0165] If a patient is diagnosed with moderate blood stasis syndrome, and the system detects that no blood-activating and stasis-removing medication order has been issued after 24 hours, a mandatory prompt will pop up on the medication order issuance interface to issue such an order to ensure the standardization of treatment. If a patient is diagnosed with blood stasis syndrome, and the system does not detect the sublingual vein purplish-darkness level assessment record for the day, a to-do task will be generated in the nurse's workbench of the nursing information system to complete the sublingual vein purplish-darkness level assessment for the patient. If a patient's qi deficiency syndrome score rises from 6 to 12 points within 8 hours, an increase of more than 5 points, the system will immediately push the warning information of increased heart failure and fall risk to the patient's bedside card and the nurse's workbench in the nursing information system. The warning information includes nursing measures suggestions such as increasing the frequency of rounds and monitoring blood pressure and blood oxygen saturation.

[0166] After nurses perform nursing interventions, they report the results in the system, forming a closed-loop management system of early warning, treatment, and feedback.

[0167] This embodiment systematically transforms the hospital information system at the data, logic, and application layers, achieving standardized collection of blood stasis syndrome elements, automated diagnosis, intelligent decision support, knowledge self-optimization, and clinical pathway quality control. Clinical trial results show that this technical solution can reduce the average TCM diagnosis time from 15 minutes to 5 minutes, improve diagnostic consistency to 88%, increase prescription efficiency by 40%, and reduce the fragmentation of TCM and Western medicine diagnosis and treatment data, providing structured, high-quality data for clinical research.

[0168] Each module of this technical solution is implemented based on existing mature technologies, exhibits good integration with hospital information systems, and does not affect the normal operation of the original systems, thus meeting the conditions for practical application. Furthermore, the solution is highly scalable; when adding other syndrome elements, only the corresponding weight rule base needs to be added to the syndrome element standard knowledge base, without reconstructing the core modules, adapting to the diagnostic and treatment needs of different syndrome elements.

[0169] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for intelligent diagnosis and decision support of TCM syndrome elements in coronary heart disease and angina pectoris based on HIS, characterized in that, The method includes the following steps: Step 1, Digital Modeling and Dynamic Rule Base Construction of Syndrome Elements: Based on the diagnostic criteria for syndrome elements of coronary heart disease and angina pectoris, the definition of the target TCM syndrome elements is deconstructed into multiple diagnostic items. A globally unique system code is assigned to each diagnostic item and a standardized data format is defined. Based on the quantitative scoring system of the diagnostic criteria, a weighted rule base stored in a machine-readable format is constructed. The weighted rule base defines the score, essential diagnostic items, diagnostic threshold, and severity grading rules for each diagnostic item. Step 2, Dynamic Structured Data Acquisition and Persistent Storage: In response to the diagnostic operation initiated by the doctor's workstation in the hospital information system, the weight rule library is called to dynamically render and generate the corresponding electronic data acquisition form interface, receive the structured data input through the form, and persistently store the structured data in the newly added TCM syndrome element fact table associated with the patient's master index and medical records in the hospital information system database. Step 3, Automated diagnosis by the intelligent computing engine: After the structured data is submitted, the intelligent computing engine service for syndrome elements deployed in the background is automatically triggered. The intelligent computing engine service for syndrome elements loads the corresponding weight rule library, verifies whether the essential diagnostic conditions are met, maps and accumulates the total score according to the weight rule library, compares the total score with the diagnostic threshold, automatically determines whether the target TCM syndrome element is established and its severity level, generates a structured diagnostic report and writes it back to the TCM syndrome element fact table; Step 4, Knowledge Graph-Driven Intelligent Decision Support: At the preset clinical business nodes of the hospital information system, the latest syndrome element diagnosis results of patients are actively obtained. Using the syndrome element diagnosis results as input, a pre-constructed TCM diagnosis and treatment knowledge graph is queried. The TCM diagnosis and treatment knowledge graph stores the semantic relationships between syndrome elements, treatment methods, prescriptions, and Chinese herbal entities. In the interface of the clinical business node, personalized treatment plans or nursing suggestions generated based on the reasoning of the TCM diagnosis and treatment knowledge graph are pushed.

2. The intelligent diagnosis and decision support method for TCM syndrome elements of coronary heart disease and angina pectoris based on HIS according to claim 1, characterized in that, In step one, the diagnostic items include core symptoms, tongue appearance, pulse appearance, objective signs, and quantitative indicators. The core symptom item is defined as a necessary item for diagnosis, and the data corresponding to the core symptom item is a Boolean value. The data for the tongue and pulse entries are classification values ​​for predefined options; The data for the quantitative indicator entries are level values.

3. The intelligent diagnosis and decision support method for TCM syndrome elements of coronary heart disease and angina pectoris based on HIS according to claim 1, characterized in that, In step three, the automated diagnostic steps of the intelligent computing engine specifically include: The intelligent calculation engine service for syndrome elements first checks whether the data of the corresponding diagnostic essential items in the submitted structured data meet the preset conditions. If not, the process is terminated. If the condition is met, all submitted diagnostic entries are traversed, and scores are mapped and accumulated according to the scoring rules defined in the weight rule library to obtain the total score. The total score is compared with the diagnostic threshold defined in the weighted rule base to determine whether the syndrome elements are valid; The severity level of the syndrome element is determined based on whether the total score falls within the severity grading interval defined in the weighted rule base.

4. The intelligent diagnosis and decision support method for TCM syndrome elements of coronary heart disease and angina pectoris based on HIS according to claim 1, characterized in that, In step four, the preset clinical business node includes a medical order issuance interface; On the medical order issuance interface, the diagnostic results of the syndrome elements are obtained through event listening. Based on the correlation between syndrome elements and treatment methods, and treatment methods and prescriptions in the TCM diagnosis and treatment knowledge graph, a personalized treatment plan that includes at least recommended prescriptions or Chinese patent medicines is generated and displayed on the side of the medical order issuance interface in the form of a prompt box or recommendation list.

5. The intelligent diagnosis and decision support method for TCM syndrome elements of coronary heart disease and angina pectoris based on HIS according to claim 1, characterized in that, The method further includes: Step 5, Clinical Data Feedback and Knowledge Evolution: Through the clinical data insight and analysis module, the structured diagnostic data accumulated in the TCM syndrome element fact table is analyzed regularly to explore the correlation patterns between symptom combinations and syndrome element diagnostic results, and knowledge update suggestions are generated. Through the knowledge rule collaborative management platform, the knowledge update suggestions are received and reviewed by experts. The approved new knowledge is then synchronously updated to the TCM diagnosis and treatment knowledge graph and the weight rule base.

6. A HIS-based intelligent diagnosis and decision support system for TCM syndrome elements in coronary heart disease and angina pectoris, applicable to the HIS-based intelligent diagnosis and decision support method for TCM syndrome elements in coronary heart disease and angina pectoris as described in any one of claims 1 to 5, characterized in that, This system integrates the following modules into the hospital information system: The syndrome element standard knowledge base module is used to store and manage a digital weighted rule base for all TCM syndrome elements; The dynamic form service engine module is used to dynamically generate a standardized data collection form interface on the front end based on the weight rule base definition in the syndrome element standard knowledge base module. The intelligent calculation engine module for syndrome elements is deployed in the form of microservices. It is used to load the weight rule library, perform automated syndrome differentiation calculations, and output structured diagnostic reports. The TCM diagnosis and treatment knowledge graph module is used to store the TCM clinical knowledge network constructed by ontology and supports semantic query and reasoning based on the diagnostic results of syndrome elements. The business scenario intelligent integration interface module is used to monitor business events in the hospital information system and push the diagnostic results output by the syndrome element intelligent calculation engine module and the recommendation information generated by the traditional Chinese medicine diagnosis and treatment knowledge graph module to the business terminal interface of the medical order system and the nursing system.

7. The intelligent diagnostic and decision support system for TCM syndrome elements of coronary heart disease and angina pectoris based on HIS according to claim 6, characterized in that, The weighted rule base stored in the syndrome element standard knowledge base module adopts JSON or XML format. The weighted rule base includes syndrome element codes, syndrome element names, definitions of essential diagnostic items, scoring rules for each diagnostic item, diagnostic thresholds, and severity grading rules.

8. The intelligent diagnosis and decision support system for TCM syndrome elements of coronary heart disease and angina pectoris based on HIS according to claim 6, characterized in that, The system also includes a clinical data insight and analysis module, which serves as a background service to periodically perform association rule analysis on the structured diagnostic data in the TCM syndrome element fact table in order to discover high-confidence symptom-syndrome element association patterns and generate structured knowledge update suggestion sheets.

9. A smart diagnostic and decision support system for TCM syndrome elements of coronary heart disease and angina pectoris based on HIS, as described in claim 6, is characterized in that, The system also includes a knowledge rule collaborative management platform, which provides a visual management interface for receiving knowledge update suggestion forms generated by the clinical data insight and analysis module, organizing expert reviews, and converting the new knowledge confirmed by the experts into new relationships or entities in the TCM diagnosis and treatment knowledge graph module, or synchronously updating the corresponding weight rule base configuration in the syndrome element standard knowledge base module.

10. A smart diagnosis and decision support system for TCM syndrome elements of coronary heart disease and angina pectoris based on HIS according to claim 6, wherein the system further includes a clinical pathway quality control rule library for syndrome elements, used to store quality control rules defined in a machine-readable format, and the business scenario intelligent integration interface module is also used to compare the patient's real-time syndrome element diagnosis results with the rules in the clinical pathway quality control rule library for syndrome elements in the clinical pathway management process of the hospital information system in real time, and to trigger mandatory prompts or warning information on the business terminal interface when deviations from the rules are detected.

Images