Patient data acquisition method and system based on mental disorder questionnaire and mapping manual

Abstract

The invention belongs to the technical field of smart medical treatment, and discloses a patient data acquisition method and system based on a mental disorder questionnaire and a mapping manual, and the acquisition method comprises the steps: obtaining individualized information, and constructing a personal graph in combination with a constructed mental disorder knowledge graph; according to principal component analysis, constructing a mental disorder assessment questionnaire for assessing symptom dimensions and a corresponding symptom dimension mapping manual; outputting a personalized questionnaire according to the personal map, the mental disorder assessment questionnaire and the symptom dimension mapping manual in combination with machine learning; and obtaining a questionnaire result, and obtaining an evaluation result according to the questionnaire result in combination with the conversion rule, thereby solving the problem that the existing mental questionnaire problem is not standard and cannot effectively interpret the symptoms of the patient.

Description

Technical Field

[0001] The present invention belongs to the field of smart medical technology, and in particular relates to a patient data collection method based on a mental disorder questionnaire and a mapping manual. Background Art

[0002] With the increasingly rapid pace of life, people face unprecedented pressures in social interactions, family life, and scientific research. These factors are continuously impacting public health and well-being, leading to a growing prevalence of psychologically induced illnesses across all social classes. Stereotypes surrounding mental health disorders can lead some patients to harbor a strong sense of stigma, leading them to seek answers online. However, search engine responses often focus on questions related to the searched domain, and the accuracy of the answers remains uncertain. Publicly available question-answering systems, such as ChatGPT, are ill-equipped to address psychiatric questions requiring specialized knowledge and high accuracy. Furthermore, due to a shortage of psychiatric medical resources, some hospitals' psychiatric outpatient clinics are underserved. First-time patients face lengthy consultations with doctors for detailed interviews and scale interpretation, while return visits require further symptom assessments. This undoubtedly increases patient stress and time, while also increasing doctors' workload and impacting the accuracy of treatment results. Therefore, we are committed to developing a tool that optimizes resource utilization, enhances doctor-patient collaboration, and comprehensively improves both patient and outpatient efficiency.

[0003] The concept of the knowledge graph (KG) was formally proposed by Google in 2012, initially with the goal of improving the accuracy of its search engine and, in turn, enhancing user experience. Knowledge graphs have been successfully applied in many fields, such as search engines, question-answering systems, and recommendation systems. Medicine, a hot topic in artificial intelligence research, is also one of the most widely used verticals for knowledge graphs, demonstrating promising applications in smart healthcare fields such as intelligent assisted medical guidance, disease risk prediction, clinical decision support, and medical Q&A. For example, Li Huafang and his team used machine learning and big data technologies to analyze and mine data, combined with the individual experience of different experts, to develop a predictive model for schizophrenia treatment outcomes. Based on this model, they constructed a knowledge graph for schizophrenia treatment outcome prediction, providing support for schizophrenia treatment decision-making. Fang Hui and his team also developed a knowledge graph for diabetes interventions, constructed a questionnaire question bank, aggregated user responses, and, based on the responses to the questionnaires, recommended diabetes interventions according to pre-set criteria. Li Zhike and his team use smartphone terminals to collect users' conversations and filter their information. They also judge users' mental health status based on their data and the constructed mental health knowledge graph, and finally transmit the analysis results to users with sub-health conditions.

[0004] However, these technologies have several limitations. First, the concepts involved in mental disorders are complex, and there may be discrepancies between the terms used by patients and their actual symptoms. However, existing knowledge graphs designed for psychiatric use fail to construct a comprehensive corpus covering all psychiatric disorders. Second, there are no quantifiable biological indicators for mental disorders. These intelligent diagnostic algorithms can only diagnose based on quantitative data provided by patients and are unable to interpret and translate the diverse symptom descriptions provided by psychiatric patients. Third, different mental illnesses require different scales, and determining the scale used and the specific symptoms corresponding to the scale items often requires extensive interpretation and assessment by physicians. Current intelligent algorithms for psychiatric use fail to generate personalized scale outputs for specific patients and fail to standardize the annotation of assessment questionnaire conclusions according to the International Classification of Diseases, 10th Revision (ICD-10). Summary of the Invention The technical problem to be solved by the present invention is to overcome the deficiencies and defects mentioned in the above background technology, and to construct a knowledge graph covering all mental illnesses by collecting structured data and unstructured data of mental symptoms in hospital systems and scientific research databases; by constructing a mental disorder questionnaire and a symptom dimension mapping manual, different questionnaire items and results are converted into ICD-10 symptom dimensions and standardized scores, and by constructing an integrated system based on intelligent reading of the knowledge graph and the symptom dimension mapping manual, personalized scales and semantic analysis results are output to patients and doctors and intelligent prompts are generated on the terminal, thereby developing a personalized patient data collection system for psychiatric outpatient clinics.

[0005] In order to solve the above technical problems, the technical solution proposed by the present invention is: A method for collecting patient data based on a mental disorder questionnaire and mapping manual, comprising the following steps: Step 1: Obtain individualized information and build a personal map based on the mental disorder knowledge map; Step 2: Based on principal component analysis, a mental disorder assessment questionnaire for assessing symptom dimensions and a corresponding symptom dimension mapping manual were constructed; Step 3: Output a personalized questionnaire based on the personal profile, mental disorder assessment questionnaire, and symptom dimension mapping manual combined with machine learning; Step 4: Receive the questionnaire results and obtain symptom classification data based on the questionnaire results and conversion rules.

[0006] In the above-mentioned personalized patient data collection method, preferably, the individualized information includes structured information and unstructured information, the structured information includes age, gender, medical diagnosis, test results, income, education level, household registration, smoking and drinking history, and occupation, and the unstructured information includes medical advice and medical record text.

[0007] More preferably, step 2 comprises: Step 201: constructing a preliminary mental disorder assessment questionnaire, establishing a mapping relationship between the questionnaire items of the preliminary mental disorder assessment questionnaire and the symptom dimensions, and obtaining a preliminary symptom dimension manual based on the mapping relationship; Step 202: Standardize the preliminary mental disorder assessment questionnaire and the preliminary symptom dimension manual, construct a data matrix, select principal components, calculate the projection matrix for the selected principal components, perform dimensionality reduction, obtain the screened questionnaire items, and obtain the mental disorder assessment questionnaire and the corresponding symptom dimension manual in combination with ICD-10.

[0008] More preferably, in step 202, the normalization process is a z-score normalization method.

[0009] More preferably, step 3 comprises: Step 301: extracting personal information based on the personal profile, and obtaining corresponding symptom dimensions based on the personal information in combination with the Mental Disorder Assessment Questionnaire and the Symptom Dimension Mapping Manual; Step 302: Use the personal information and the corresponding symptom dimensions as a data set, input them into machine learning training and verification, and output a personalized questionnaire.

[0010] More preferably, in step 302, the data set is processed by synthesizing a small number of oversampled data and then put into machine learning training and validation; The algorithm used in the machine learning is a random forest algorithm based on the mapping relationship between questionnaire items and symptom dimensions and optimized through grid search.

[0011] More preferably, when the questionnaire result is a numerical value, obtaining the evaluation result based on the questionnaire result in combination with the conversion rule includes: When the questionnaire result is a score of 0-4, it is directly output as the evaluation result; Among them, the 0-4 score level means that the score of the questionnaire result is in the range of 0-4, 4 is the highest score, and 0 is the lowest score; When the questionnaire results are of other score levels, the evaluation results are output in combination with the linear mapping rule; The linear mapping rule is expressed as: ; in, Indicates that the evaluation has results; Indicates the questionnaire result score; and Respectively represent the upper limit and lower limit of the target score; and They represent the upper limit and lower limit of the original score respectively.

[0012] More preferably, when the questionnaire result is of a language type, obtaining the evaluation result based on the questionnaire result in combination with the conversion rule includes: obtaining key information based on the natural language recognition questionnaire result, and obtaining the evaluation result based on the key information in combination with the scoring rule; The scoring rule is: 0 points for no symptoms, 1 point for mild symptoms, 2 points for moderate symptoms, 3 points for obvious symptoms, and 4 points for severe symptoms.

[0013] Based on a general inventive concept, the present invention also provides a patient data collection system based on a mental disorder questionnaire and mapping manual, comprising a symptom knowledge graph module, a questionnaire and symptom mapping manual module, a machine learning output personalized questionnaire module, and a questionnaire analysis module. The symptom knowledge graph module stores a mental symptom knowledge graph constructed based on public medical record data. The symptom knowledge graph module is used to combine the acquired individualized information with the mental disorder knowledge graph to construct a personal graph; The questionnaire and symptom mapping manual module is used to construct a mental disorder assessment questionnaire and a disorder dimension mapping manual based on principal component analysis; The machine learning output personalized questionnaire module is used to call the personal atlas, mental disorder assessment questionnaire, and disorder dimension mapping manual and combine them with machine learning to output a personalized questionnaire; The questionnaire analysis module is used to obtain questionnaire results and obtain evaluation results based on the questionnaire result type and different conversion rules.

[0014] The above-mentioned patient data collection system preferably further includes a voice collection module and a semantic conversion module; The voice collection module is used to collect the patient's voice data and transmit the patient's voice data to the semantic conversion module; The semantic conversion module is used to combine the medical consultation voice data with semantic analysis to obtain medical consultation semantic text, and transmit the medical consultation semantic text to the symptom knowledge graph module; The symptom knowledge graph module receives the medical consultation semantic text as individualized information and combines it with the mental disorder knowledge graph to construct a personal graph.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention gathers information resources on mental illness symptoms and, in combination with objective medical facts, constructs a medical knowledge graph centered on symptom knowledge. This includes a mental symptom knowledge graph ontology model layer and a mental symptom knowledge graph data layer that provides actual data. Effective artificial intelligence technology is used to graphically, semantically, and structurally represent professional knowledge on mental illnesses, enabling computer recognition and calculation. 2. Using patient information in the mental symptom knowledge graph data layer and symptom dimension information in the mapping manual as training data, based on machine learning technology, through model training and verification, we obtain the assessment questionnaire set corresponding to the symptom dimension that best matches the patient information, thereby achieving personalized push of patient assessment questionnaires and improving medical treatment efficiency; 3. The personalized questionnaires obtained through machine learning are more targeted and can collect patients' digital symptom results more quickly and in a standardized manner, effectively reducing the workload of doctors. Rapid classification and grading based on patient data are beneficial for quickly retrieving or calling data information in the later stage of diagnosis, which is beneficial to improving the accuracy of subsequent diagnosis.

[0016] 4. By calling the Mental Disorder Assessment Questionnaire and Symptom Dimension Mapping Manual, combined with the rule conversion of questionnaire scoring, different questionnaire items and results are converted into a unified symptom dimension and standardized score, and the various symptom dimensions and scores are summarized to improve the standardization of the data, which is also conducive to data summary and utilization. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 is a flow chart of the patient data collection method of the present invention; Figure 2 is a schematic diagram of the construction mechanism of the patient data acquisition system of the present invention; Figure 3 is a flow chart of the use of the patient data acquisition system of the present invention; Figure 4 It is a schematic diagram of the component modules of the patient data acquisition system of the present invention. DETAILED DESCRIPTION

[0019] To facilitate understanding of the present invention, the present invention will be described in more comprehensive and detailed form below in conjunction with the accompanying drawings and preferred embodiments. However, the protection scope of the present invention is not limited to the following specific embodiments.

[0020] Unless otherwise defined, all technical terms used hereinafter have the same meanings as those generally understood by those skilled in the art. The technical terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the scope of protection of the present invention.

[0021] Unless otherwise specified, various raw materials, reagents, instruments and equipment used in the present invention can be purchased from the market or prepared by existing methods.

[0022] A knowledge graph is a semantic network structure that represents relationships between entities through nodes and edges, providing a framework for integrating information from multiple sources. In the assessment of mental disorders, knowledge graphs integrate multiple aspects of information, including patient medical history, genetic information, symptoms, and medical literature, to help doctors gain a comprehensive understanding of the patient's condition and facilitate the development of personalized diagnosis and treatment plans. They help quickly discover connections between factors, support doctors in identifying potential causes, improve the precision of treatment, and provide a reference for predicting potential risks for patients. Knowledge graphs provide deeper and more comprehensive data analysis and applications for the mental health field, and are expected to play a significant role in improving the assessment and treatment of mental disorders.

[0023] Example: See Figure 1 , a patient data collection method based on the mental disorder questionnaire and mapping manual, comprising the following steps: S1. Construction of the knowledge graph of mental symptoms Gather information resources on mental illness symptoms and combine them with objective medical facts to construct a medical knowledge graph with symptom knowledge as the core, namely the mental symptom knowledge graph ontology model layer and the mental symptom knowledge graph data layer that provides actual data; S11. Collect patient information extensively. The included patients mainly come from information collected from the hospital medical record system and the individualized information corresponding to the patients in the medical record system. The source of individualized information includes scientific research information surveys in which patients participate. All information includes structured information and unstructured information. The former includes the patient's age, gender, disease diagnosis, laboratory test results, questionnaire assessment results, etc. from the electronic medical record system, as well as disease risk factor information collected through scientific research projects, such as family income, education level, household registration, smoking and drinking history, occupation, etc. The latter is based on the characteristics of psychiatric diagnosis and treatment and only includes text information such as doctor's orders and medical records; S12. Combine objective medical facts to construct a knowledge graph of mental symptoms and establish conceptual entity relationship entities of mental symptoms, specifically; Conceptual entities can be symptom types, patients, questionnaires, etc. Relational entities can be the frequency, duration, location, nature, severity, diurnal characteristics, seasonal characteristics, triggering factors, relief factors, etc. of symptoms; Attributes can be information corresponding to relational entities, for example, day and night characteristics can be light in the morning and heavy in the evening.

[0024] S13. Integrate individual structured data and unstructured data to obtain a personal graph.

[0025] After preprocessing the patient's structured and unstructured data, conceptual entities, relationships, and attribute knowledge units are extracted from the data.

[0026] Through machine learning, the conceptual entities, relational entities, and attributes of mental symptoms are treated as a binary classification problem. Using supervised learning techniques, the model is trained using pre-annotated data, and the unannotated data is aligned, fused, and optimized to form a personal atlas.

[0027] S2. Construction of a Mental Disorder Assessment Questionnaire and Symptom Dimension Mapping Manual The symptom dimensions assessed by mental disorder questionnaires are usually reflected by a single item or multiple items, and the correspondence between questionnaire items and symptom dimensions is defined as "mapping"; S21. The establishment of a manual for mapping mental disorder assessment questionnaires and symptom dimensions will involve mapping and annotating the symptom dimensions corresponding to commonly used mental disorder questionnaire items through literature research and expert discussions. S22. For questionnaires and items with a large number of items and complex relationships between items, which makes it difficult to implement symptom dimension labeling in step S11, use principal component analysis (PCA) to reduce the dimension of the questionnaire items and extract characteristic components; S22.1. Data preparation and standardization: Based on the knowledge graph data layer of mental symptoms, extract symptom assessment information of patients with different diseases according to disease type, and use the z-score standardization method or scale the data to the same range; S22.2. Construct a data matrix: Organize the data into a matrix where rows represent samples and columns represent features. Calculate the covariance matrix, eigenvalues, and eigenvectors: Calculate the covariance matrix between features to describe the correlation or linear relationship between features; Perform eigenvalue decomposition on the covariance matrix to obtain eigenvalues and corresponding eigenvectors. The eigenvectors represent the main directions or main components in the data; S22.3. Select principal components: Select the number of principal components to retain based on the size of the eigenvalues. Select the first few principal components with larger eigenvalues because they contain the most important variation information in the data. S22.4. Calculate projection matrix: Use the selected principal components to construct a projection matrix and project the original data into the new principal component space; S22.5, Dimensionality reduction: Reduce the original data to a new space of lower dimensionality by retaining the principal components; S23, control and ICD-10 diagnostic criteria were used to standardize symptom dimension annotations for the newly divided dimensions.

[0028] S3. Personalized evaluation questionnaire output module Using patient information in the mental symptom knowledge graph data layer and symptom dimension information in the mapping manual as training data, based on machine learning technology, through model training and verification, we obtain the assessment questionnaire set corresponding to the symptom dimension that best matches the patient information, thereby achieving personalized push of patient assessment questionnaires; S31. Extracting the patient's personalized data as feature information from the mental symptom knowledge graph data layer; S32. Use Scikit-Learning as a machine learning reference framework; S33. Considering the low prevalence of some mental disorders, the data is somewhat unbalanced, so the synthetic minority oversampling technique (SMOTE) is used to process the dataset. SMOTE avoids overfitting by synthesizing new samples along a straight line between a given sample and one of its neighboring samples; S34. Use grid search to perform hyperparameter tuning to optimize model performance; S35. Use the random forest classification algorithm to build a model; use cross-validation and machine learning models to predict the symptom dimensions and calculate the weight of each feature using relevant features; By using the random forest method, a large number of trees are built in the sample, and then an aggregate tree is generated by averaging all the trees.

[0029] S4. Questionnaire result analysis module This module mainly uses S1's Mental Disorder Questionnaire and Symptom Dimension Mapping Manual, combines it with questionnaire scoring rule conversion, converts different questionnaire items and results into a unified symptom dimension and standardized score, summarizes each symptom dimension and score result, and outputs an assessment report; S41. Call the Mental Disorder Questionnaire and Symptom Dimension Mapping Manual in S2 and mark the symptoms corresponding to the questionnaire items; S42. Different conversion rules are called according to the questionnaire score data type to achieve questionnaire score standardization; (1) When the score is a numerical variable, for the numerical grade results of 0-4 points, the original score result value is directly used as the final result value; for the results of 0-10 points or other numerical grades, the linear mapping rule is used, and its principle is conversion score = original score × (target score upper limit - target score lower limit) / (original score upper limit - original score lower limit) + target score lower limit, where the symptom dimension corresponding to the questionnaire item is based on the ICD-10 regulations. (2) When the score is a descriptive term, such as "depressed" or "elated", combined with Natural Language Processing (NLP) technology, a branch of artificial intelligence that focuses on enabling computers to understand, interpret, and operate human language, by identifying keyword information in descriptive terms, a standard scoring scale of 0-4 points is defined as follows: 0 is no symptoms (none, never, almost never, etc.), 1 is mild symptoms (sometimes, occasionally, does not affect life, etc.), 2 is moderate symptoms (about half of the time, sometimes can be overcome, has a certain impact on work / study / social life, etc.), 3 is obvious symptoms (most of the time, difficult to relieve on their own, obviously affects work / life / social life, feels painful, etc.), 4 is severe symptoms (always, almost every day, difficult to take care of oneself, subjective extreme pain); The rule system offers flexible configuration options, allowing users to customize the conversion of symptom severity levels as needed. Users can adjust and optimize the parameters and rules of the rule conversion based on the needs of specific assessment questionnaires.

[0030] For each specific questionnaire item and result, this system works in-depth with experts in the psychiatric field to guide the implementation of conversion rules, design personalized mapping radiation for questionnaire items, and record them in the system to deal with situations where existing conversion rules cannot correctly reflect symptom severity and clinical usability.

[0031] This system has the ability to dynamically update the knowledge graph, that is, the system can timely update and expand the transformation rules as new evaluation questionnaires are introduced and domain knowledge advances to adapt to the ever-changing needs of evaluation questionnaires.

[0032] S43. For the scoring results, the user can mark the current symptom item as "intervention required". If the score is 2 points higher than the intervention threshold, the current symptom item is marked as "intervention required"; if the score is lower than the intervention threshold, the current symptom item is marked as "no intervention required"; S44. When summarizing all questionnaire items and results, the symptom items and results classified into the same category will be averaged and then compared with the intervention threshold to output the intervention markers and clinical significance of all symptom results.

[0033] See Figure 2-4 , an embodiment of the present invention also provides a patient data collection system based on a mental disorder questionnaire and mapping manual, including a symptom knowledge graph module, a questionnaire and symptom mapping manual module, a voice collection module, and a semantic conversion module; Machine learning outputs a personalized questionnaire module and a questionnaire analysis module, and the symptom knowledge graph module stores a mental symptom knowledge graph built based on public medical record data; The voice collection module is used to collect the patient's voice data and transmit the patient's voice data to the semantic conversion module; The semantic conversion module is used to combine the medical consultation voice data with semantic analysis to obtain the medical consultation semantic text, and transmit the medical consultation semantic text to the symptom knowledge graph module; The symptom knowledge graph module is used to receive the semantic text of the medical consultation as individualized information and build a personal graph based on the mental disorder knowledge graph; Questionnaire and symptom mapping manual module, used to construct a mental disorder assessment questionnaire and disorder dimension mapping manual based on principal component analysis; The machine learning output personalized questionnaire module is used to call the personal map, mental disorder assessment questionnaire, and disorder dimension mapping manual and combine them with machine learning to output personalized questionnaires; The questionnaire analysis module is used to obtain questionnaire results and obtain evaluation results based on the questionnaire result type and different conversion rules.

[0034] Experimental case 1: Obtain patient information to obtain a patient model, including information such as gender, age, perinatal experience, BMI, lifestyle, and economic status. The patient information obtained is as follows: female, 25 years old, perinatal, BMI of 16, a lifestyle that likes to stay up late, and low income; visiting a psychiatric department, and pushing the PHQ-9 questionnaire (The Patient Health Questionnaire-9) to the patient. Taking the first item of the questionnaire as an example, that is, "I feel no interest or pleasure in doing anything", by calling the Mental Disorder Questionnaire and Symptom Dimension Mapping Manual, the symptom induced by this item is "loss of interest", and the item result is "3 points"; the item results of the assessment questionnaire are converted into the result value of the current assessment item consistent with the standard scoring level, and then compared with the current symptom intervention critical score of 2 points, and this symptom item is marked as "need / no intervention"; In this embodiment, the first item of the PHQ-9 has a score of "3" and is a continuous variable. The scoring level is consistent with the data type of the standard score. However, if the scoring levels are inconsistent, a linear mapping rule is used. The principle is: conversion score = original score × (target range upper limit - target range lower limit) / (original range upper limit - original range lower limit) + target range lower limit; the questionnaire item result value is converted into the result value of the current questionnaire item consistent with the standard score, that is, 3*(4-0) / (3-0)+0=4.

[0035] In this embodiment, the result value of the first item of PHQ-9 is 4, which is greater than the critical score. The clinical significance of the symptom result is "the patient's loss of interest is severe, affecting daily life and requiring timely intervention. This symptom is common in affective disorders, such as depression, and needs to be further analyzed in combination with other symptoms."

[0036] Repeat steps S22 to S24 to analyze each questionnaire item and result of all evaluation questionnaires in turn, obtain the "intervention required" mark and clinical significance analysis of each questionnaire item and result; summarize the "intervention required" mark and clinical significance analysis of all questionnaire items and results, and output the test report.

[0037] In this experimental case, the patient was assessed with the PHQ-9, and all results were summarized and reported in a spreadsheet. The analysis report included the overall questionnaire assessment results and the analysis results of the questionnaire items (eliciting symptoms and intervention markers). Clicking "Action" would output the results and prompt a report on the analytical significance of the corresponding symptoms or overall questionnaire score.

[0038] An example summary of all the results is as follows: This patient is a perinatal woman at high risk for depression, with a PHQ-9 score of 23, suggesting possible severe depression. The patient has significant symptoms of "decreased interest," "depressed mood," "poor sleep," "decreased energy," "self-blame," "decreased concentration," and "self-harm thoughts," requiring further questioning and evaluation by the doctor, and appropriate intervention measures. In addition, the patient has the following risk factors that require attention: low income, and attention should be paid to the perinatal status to prevent adverse perinatal events.

[0039] Experimental case 2: Obtain patient information to create a patient model, including gender, age, BMI, lifestyle, appetite, sleep, and physical symptoms. The patient information obtained is as follows: a 30-year-old female with a BMI of 17.2, excessively worried about work and family, poor appetite, difficulty falling asleep, tension headaches, palpitations, and upper abdominal discomfort. She visited a psychiatric clinic and was given the Generalized Anxiety Disorder-7 (GAD-7) questionnaire. For example, the first item in the questionnaire, "feeling uneasy, worried, and irritable," was used to map the mental disorder questionnaire to the symptom dimension. This item resulted in the symptom "panic," resulting in a score of "3." The results of the assessment questionnaire items are converted into the result values of the current assessment items consistent with the standard scoring level, and then compared with the current symptom intervention critical score of 2 points, and this symptom item is marked as "need / no need intervention"; In this embodiment, the first item of the GAD-7 has a score of "3," which is a continuous variable. The scoring level is consistent with the data type of the standard score. However, if the scoring levels are inconsistent, a linear mapping rule is used. The principle is that the conversion score = original score × (target range upper limit - target range lower limit) / (original range upper limit - original range lower limit) + target range lower limit. The questionnaire item result value is converted into the result value of the current questionnaire item consistent with the standard score, that is, 3*(4-0) / (3-0)+0=4.

[0040] In this embodiment, the result value of the first item of GAD-7 is 4, which is greater than the critical score. The clinical significance of the symptom result is that "the patient has been troubled for a long time by worrying about bad things that may happen in the future, feeling uneasy, and affecting daily life. Timely intervention is needed. This symptom is common in generalized anxiety disorder and needs to be further analyzed in combination with other symptoms."

[0041] Repeat steps S22 to S24 to analyze each questionnaire item and result of all assessment questionnaires in turn, and obtain the "need intervention" mark and clinical significance analysis of each questionnaire item and result; Summarize the "intervention required" mark and clinical significance analysis of all questionnaire items and results, and output the test report.

[0042] In this experimental case, the patient was assessed for GAD-7. A summary of all results and a detailed report are shown in Table 1. This report will be generated in a spreadsheet format. The analysis report includes the overall questionnaire assessment results and the analysis results of the questionnaire items (eliciting symptoms and intervention markers). Clicking "Action" will output the results, which will provide a report on the analytical significance of the corresponding symptoms or overall questionnaire score. An example summary of all results is as follows: This patient is a middle-aged woman at high risk for anxiety. Her GAD-7 score is 13, suggesting the possibility of moderate anxiety disorder. She also has significant symptoms of "panic" and "motor anxiety," requiring further questioning and evaluation by a physician, along with appropriate intervention measures.

[0043] Table 1 Summary of analysis results of patient assessment questionnaire

[0044] Experimental case three: Patient information was obtained to create a patient model, including gender, age, lifestyle, social life, and sleep patterns. The patient information obtained was as follows: a 32-year-old male who enjoys spending and gift-giving, strikes up conversations with unfamiliar women, easily argues with family members, sleeps less but remains energetic during the day, and seeks psychiatric care. The patient was given the Young Mania Rating Scale (YMRS) questionnaire. For example, the fifth item on the questionnaire, "Subjectively Feeling Irritable," was found by using the Mental Disorders Questionnaire and Symptom Dimension Mapping Manual. The symptom elicited by this item was "Irritability," resulting in a score of "2." The results of the assessment questionnaire items are converted into the result values of the current assessment items consistent with the standard scoring level, and then compared with the current symptom intervention critical score of 2 points, and this symptom item is marked as "need / no need intervention"; In this embodiment, the result of the fifth item of the YMRS is "2 points", which is a continuous variable. The scoring level is consistent with the data type of the standard scoring, but the scoring levels are inconsistent. A linear mapping rule is used, the principle of which is conversion score = original score × (target range upper limit - target range lower limit) / (original range upper limit - original range lower limit) + target range lower limit, to convert the questionnaire item result value into the result value of the current questionnaire item consistent with the standard score, that is, 2*(4-0) / (8-0)+0=1.

[0045] In this embodiment, the result value of the first item of YMRS is 1, which is less than the critical score. The clinical significance of the symptom result is that "the patient's irritability symptoms are not obvious and no treatment is needed for the time being."

[0046] Repeat steps S22 to S24 to analyze each questionnaire item and result of all assessment questionnaires in turn, and obtain the "need intervention" mark and clinical significance analysis of each questionnaire item and result; Summarize the "intervention required" mark and clinical significance analysis of all questionnaire items and results, and output the test report.

[0047] In this experimental case, the patient underwent the YMRS assessment. A summary of all results and a detailed report are shown in Table 2. This report will be generated in a spreadsheet format. The analysis report includes the overall questionnaire assessment results and the results of the questionnaire item analysis (eliciting symptoms and intervention markers). Clicking "Action" will output the results, which will provide a report on the analytical significance of the corresponding symptom or overall questionnaire score. An example summary of all results is as follows: This patient is a middle-aged male with a YMRS score of 26, suggesting possible bipolar disorder; the patient's "elevated mood" and "increased activity and energy" symptoms are significant, requiring further questioning and evaluation by a physician, and appropriate intervention measures.

[0048] Table 2 Summary of analysis results of patient assessment questionnaire

[0049] In summary, the present invention constructs a knowledge graph of mental symptoms, which maps, semanticizes, and structure clinical expertise. It also constructs a mapping manual for mental disorder questionnaire items and symptom dimensions by combining literature research, expert discussions, and statistical analysis. Using machine learning, it matches the patient's symptom dimensions with their individualized information. Combining the mental questionnaire items with the symptom mapping manual helps identify the most appropriate assessment questionnaire for each patient, facilitating personalized questionnaire delivery for each patient. Based on this, a set of mental questionnaire item analysis methods was developed, which achieves normalization and standardization of questionnaire scores and data standardization. Through the use of diverse information, effective interaction between doctors and patients and flexible data management are achieved.

Claims

1. A method for collecting patient data based on a mental disorder questionnaire and mapping manual, characterized in that: The steps include: Step 1: Obtain individualized information and build a personal map based on the mental disorder knowledge map; Step 2: Based on principal component analysis, a mental disorder assessment questionnaire for assessing symptom dimensions and a corresponding symptom dimension mapping manual were constructed; Step 3: Output a personalized questionnaire based on the personal profile, mental disorder assessment questionnaire, and symptom dimension mapping manual combined with machine learning; Step 4: Receive the questionnaire results and obtain symptom classification data based on the questionnaire results and conversion rules.

2. The patient data collection method based on the mental disorder questionnaire and mapping manual according to claim 1 is characterized in that: The individualized information includes structured information and unstructured information. The structured information includes age, gender, medical diagnosis, test results, income, education level, household registration, smoking and drinking history, and occupation. The unstructured information includes medical advice and medical record text.

3. The patient data collection method based on the mental disorder questionnaire and mapping manual according to claim 1, characterized in that: Step 2 includes: Step 201: constructing a preliminary mental disorder assessment questionnaire, establishing a mapping relationship between the questionnaire items of the preliminary mental disorder assessment questionnaire and the symptom dimensions, and obtaining a preliminary symptom dimension manual based on the mapping relationship; Step 202: Standardize the preliminary mental disorder assessment questionnaire and the preliminary symptom dimension manual, construct a data matrix, select principal components, calculate the projection matrix for the selected principal components, perform dimensionality reduction, obtain the screened questionnaire items, and obtain the mental disorder assessment questionnaire and the corresponding symptom dimension manual in combination with ICD-10.

4. The patient data collection method based on the mental disorder questionnaire and mapping manual according to claim 3 is characterized in that: In step 202, the normalization process is a z-score normalization method.

5. The patient data collection method based on the mental disorder questionnaire and mapping manual according to claim 3, characterized in that: Step 3 includes: Step 301: extracting personal information based on the personal profile, and obtaining corresponding symptom dimensions based on the personal information in combination with the Mental Disorder Assessment Questionnaire and the Symptom Dimension Mapping Manual; Step 302: Use the personal information and the corresponding symptom dimensions as a data set, input them into machine learning training and verification, and output a personalized questionnaire.

6. The method for collecting patient data based on the mental disorder questionnaire and mapping manual according to claim 5, characterized in that: In step 302, the dataset is processed by synthesizing a small number of oversampled data and then put into machine learning training and validation; The algorithm used in the machine learning is a random forest algorithm based on the mapping relationship between questionnaire items and symptom dimensions and optimized through grid search.

7. The method for collecting patient data based on the mental disorder questionnaire and mapping manual according to any one of claims 1 to 6, characterized in that: When the questionnaire result is a numerical value, obtaining the evaluation result based on the questionnaire result combined with the conversion rules includes: When the questionnaire result is a score of 0-4, it is directly output as the evaluation result; When the questionnaire results are of other score levels, the evaluation results are output in combination with the linear mapping rule; The linear mapping rule is expressed as: ； in, Indicates that the evaluation has results; Indicates the questionnaire result score; and Respectively represent the upper limit and lower limit of the target score; and They represent the upper limit and lower limit of the original score respectively.

8. The method for collecting patient data based on the mental disorder questionnaire and mapping manual according to any one of claims 1 to 6, characterized in that: When the questionnaire result is of a language type, obtaining the evaluation result based on the questionnaire result in combination with the conversion rule includes: obtaining key information based on the natural language recognition questionnaire result, and obtaining the evaluation result based on the key information in combination with the scoring rule; The scoring rule is: 0 points for no symptoms, 1 point for mild symptoms, 2 points for moderate symptoms, 3 points for obvious symptoms, and 4 points for severe symptoms.

9. A patient data collection system based on a mental disorder questionnaire and mapping manual, characterized in that: It includes a symptom knowledge graph module, a questionnaire and symptom mapping manual module, a machine learning output personalized questionnaire module, and a questionnaire analysis module. The symptom knowledge graph module stores a mental symptom knowledge graph built based on public medical record data; The symptom knowledge graph module is used to combine the acquired individualized information with the mental disorder knowledge graph to construct a personal graph; The questionnaire and symptom mapping manual module is used to construct a mental disorder assessment questionnaire and a disorder dimension mapping manual based on principal component analysis; The machine learning output personalized questionnaire module is used to call the personal atlas, mental disorder assessment questionnaire, and disorder dimension mapping manual and combine them with machine learning to output a personalized questionnaire; The questionnaire analysis module is used to receive questionnaire results and obtain evaluation results based on the questionnaire result type and different conversion rules.

10. The patient data collection system based on the mental disorder questionnaire and mapping manual according to claim 9, characterized in that: It also includes a voice acquisition module and a semantic conversion module; The voice collection module is used to collect the patient's voice data and transmit the patient's voice data to the semantic conversion module; The semantic conversion module is used to combine the medical consultation voice data with semantic analysis to obtain medical consultation semantic text, and transmit the medical consultation semantic text to the symptom knowledge graph module; The symptom knowledge graph module receives the medical consultation semantic text as individualized information and combines it with the mental disorder knowledge graph to construct a personal graph.

Images