An etiology screening and auxiliary decision-making method and system based on a structured electronic questionnaire

Abstract

The application discloses a kind of based on structured electronic questionnaire etiology screening and auxiliary decision-making method and system, it is related to digital health assessment and clinical decision support technical field, including obtaining the disease characteristics of patient, utilize disease characteristics to enter corresponding etiology diagnosis question group etiology identification, based on feature mapping rule etiology diagnosis question group carries out etiology score calculation, obtains candidate etiology score set, based on etiology threshold etiology score set is judged and the etiology that meets requirement is output, obtains etiology preliminary judgment result;According to etiology preliminary judgment result obtains final auxiliary decision-making;Final auxiliary decision-making is used to assist doctor to make clinical diagnosis and disposal.The application realizes disease history standardization, quantifiable, traceable by structured questionnaire and logic branch, reduces the omission and repeated inquiry situation of patient information acquisition process, improves patient treatment efficiency, reduces the work intensity of doctor.

Description

Technical Field

[0001] This invention relates to the field of digital health assessment and clinical decision support technology, and in particular to a method and system for etiological screening and decision support based on structured electronic questionnaires. Background Technology

[0002] Cough is one of the most common medical symptom presented in children. Clinically, a cough lasting ≥4 weeks is generally defined as chronic cough; some guidelines use chest X-rays and other examinations during assessment to rule out serious organic lesions. The causes of chronic cough in children are complex and phenotypically overlapping. Common causes include post-infection cough (PIC), upper airway cough syndrome (UACS), cough variant asthma (CVA), and protracted bacterial bronchitis (PBB). Gastroesophageal reflux-related cough (GERC), foreign body aspiration, airway anatomical abnormalities, and immunodeficiency should also be considered.

[0003] In actual outpatient and primary healthcare settings, medical history collection relies on doctors' experience and the time available for on-site consultation. Parents' descriptions of the nature, causes, timing, and accompanying symptoms of coughs are often subjective and fragmented, leading to the omission of key information, resulting in misdiagnosis, missed diagnosis, or unnecessary examinations and inappropriate medication (such as unnecessary use of antibiotics).

[0004] While domestic and international guidelines and consensus statements provide treatment pathways, there is a lack of standardized tools that are "parent-oriented, can collect structured data, can automatically generate etiology prompts, can be quickly accessed by doctors, and can be verified through follow-up and closed-loop processes." Existing general questionnaires or educational materials often fail to provide logical branching, etiology mapping, and clinical decision-making prompts, and also lack comparative verification and traceable data recording in real outpatient procedures.

[0005] Therefore, proposing a method and system for etiological screening and decision support based on structured electronic questionnaires to address the difficulties of existing technologies is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0006] In view of this, the present invention provides a method and system for etiology screening and decision support based on structured electronic questionnaires. Through structured questionnaires and interpretable rules, etiology prompts can be automatically generated based on the questionnaire content, supporting patients' preliminary self-testing and doctors' decision support. This helps to reduce examinations and medications in the disease diagnosis process, improves consultation efficiency, and reduces the workload of doctors.

[0007] To achieve the above objectives, the present invention adopts the following technical solution: A method for etiological screening and decision support based on structured electronic questionnaires, comprising: S1 data collection steps: Obtain the patient's disease characteristics through the completed questionnaire; S2 identification steps: Based on the questionnaire jump rules, the etiology is identified using disease characteristics, and the corresponding etiology diagnosis question group is entered; the etiology diagnosis question group includes various etiology diagnosis questions corresponding to the disease characteristics; S3 Calculation Steps: Based on the feature mapping rules, calculate the etiology scores for the etiology diagnosis question set to obtain a candidate etiology score set; the candidate etiology score set includes etiology scores that correspond one-to-one with the etiology. S4 Judgment Step: Based on the etiology threshold, the candidate etiology score set is judged, and the etiology that meets the requirements is output to obtain the preliminary etiology judgment result; S5 auxiliary step: Based on the preliminary judgment of the etiology, combined with the patient's physical examination and auxiliary examinations, a final auxiliary decision is made.

[0008] In the above method, optionally, in S1, the disease characteristic is cough symptoms in children.

[0009] In the above methods, optionally, in S2, the etiological diagnosis question group includes: cough variant asthma diagnosis question, allergic cough diagnosis question, asthma diagnosis question, upper airway cough syndrome diagnosis question, gastroesophageal reflux-related cough diagnosis question, post-infectious cough diagnosis question, psychogenic / tic-related cough diagnosis question, pertussis diagnosis question, pulmonary tuberculosis diagnosis question, persistent bacterial bronchitis diagnosis question, bronchiectasis diagnosis question, persistent pneumonia and chronic pneumonia diagnosis question, cystic fibrosis diagnosis question, congenital laryngotracheomalacia diagnosis question, primary ciliary dyskinesia diagnosis question, and drug-induced cough diagnosis question.

[0010] In the above method, optionally, the feature mapping rule in S3 is the 6T-6S feature mapping rule; the 6T-6S features include: 6T entry features and 6S entry features; The 6T features include onset time, triggering and aggravating factors, disease course, cough type, accompanying symptoms, and previous treatment response; the 6S features include sneezing and allergies, swallowing and bowel movements, sleep, movement, severity-persistence-rare-recurrence, and voice characteristics.

[0011] In the above method, the optional 6T-6S feature mapping rule determination process includes: Etiological clues were extracted based on guidelines / consensus and clinical diagnostic criteria and converted into cough characteristic options; Through discussions among clinical experts, cough feature options were categorized, merged, and redundancy removed to obtain the basic information dimension and the quick prompt clue dimension for cough. The 6T+6S feature mapping rule is determined based on the dimensions of basic cough information and quick cueing.

[0012] In the above method, optionally, the questionnaire in S1 is prepared based on cough characteristic options.

[0013] In the above method, optionally, the formula for calculating the candidate cause score in S3 is: ; in, j Indicates candidate etiology j , S jk Indicates candidate etiology j The score, I ( option jk ) indicates whether to select.

[0014] The above methods may optionally include: S6 scoring steps: Obtain and record the confidence score and help score of the final auxiliary decision; the confidence score is used to characterize the doctor's degree of acceptance of the final auxiliary decision, and the help score is used to characterize the reliability of the final auxiliary decision.

[0015] The above methods may optionally include: S7 Validation Steps: Follow up and evaluate patients based on SOP procedures to obtain optimized rule thresholds and weight annotation data.

[0016] A system for etiology screening and decision support based on structured electronic questionnaires, used to implement an etiology screening and decision support method based on structured electronic questionnaires as described above, including a patient-side questionnaire module, a logic branching and feature extraction module, an etiology inference and scoring engine, a doctor-side display and decision support module, a follow-up and verification module, and a data storage and security module connected in sequence. The patient-side questionnaire module allows patients or their families to complete questionnaires on mobile devices via QR codes or links, thereby obtaining information about the patient's disease characteristics. The logic branching and feature extraction module is used to identify the cause of disease based on disease characteristics, trigger branch questions, and enter the corresponding etiology diagnosis question group. The etiology inference and scoring engine is used to calculate etiology scores for etiology diagnosis questions to obtain a set of candidate etiology scores. Based on etiology thresholds, the engine judges the candidate etiology scores and outputs the etiology that meets the requirements, thus obtaining a preliminary etiology judgment result. The doctor's side display and decision support module is used to present the questionnaire answers and preliminary diagnosis results, and record the doctor's diagnosis results, whether the questionnaire was referenced, the helpness score and the confidence score. The follow-up and verification module is used to conduct follow-up assessments according to the SOP procedure based on the preset follow-up cycle. The data storage and security module is used for data anonymization and encrypted storage, and provides access control and log traceability.

[0017] As can be seen from the above technical solution, compared with the prior art, the present invention provides a method and system for etiological screening and auxiliary decision-making based on structured electronic questionnaires, which has the following beneficial effects: (1) By using structured questionnaires and logical branches, the key medical history of chronic cough is collected in a "standardized, quantifiable and traceable" manner, reducing information omissions and repeated consultations; (2) It realizes auxiliary decision-making based on interpretable rules and scoring engine, and automatically generates etiology prompts based on the questionnaire content, which can effectively avoid information omission and reduce the workload of doctors. (3) It supports two scenarios: self-testing by patients or their families and doctor assistance, which improves the timeliness and efficiency of medical treatment to a certain extent, and achieves assessable clinical implementation by recording the degree of acceptance and assistance. (4) By embedding a follow-up closed loop, the initial diagnosis is verified by subsequent treatment response and outcome, providing a real data basis for the continuous optimization of etiology screening and auxiliary decision-making.

[0018] (5) It helps reduce unnecessary examinations and medications for patients, and promotes hierarchical medical treatment and precise intervention in disease diagnosis and treatment. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0020] Figure 1 This is a flowchart of a method for etiological screening and decision support based on a structured electronic questionnaire disclosed in this invention; Figure 2 This is a schematic diagram of the data collection process in an outpatient setting as disclosed in an embodiment of the present invention; Figure 3 This is a schematic diagram of the follow-up and closed-loop verification process disclosed in an embodiment of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] In this application, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0023] Reference Figure 1 This invention discloses a method for etiological screening and decision support based on a structured electronic questionnaire, comprising: S1 data collection steps: Obtain the patient's disease characteristics through the completed questionnaire; S2 identification steps: Based on the questionnaire jump rules, the etiology is identified using disease characteristics, and the corresponding etiology diagnosis question group is entered; the etiology diagnosis question group includes various etiology diagnosis questions corresponding to the disease characteristics; S3 Calculation Steps: Based on the feature mapping rules, calculate the etiology scores for the etiology diagnosis question set to obtain a candidate etiology score set; the candidate etiology score set includes etiology scores that correspond one-to-one with the etiology. S4 Judgment Step: Based on the etiology threshold, the candidate etiology score set is judged, and the etiology that meets the requirements is output to obtain the preliminary etiology judgment result; S5 auxiliary step: Based on the preliminary judgment of the etiology, combined with the patient's physical examination and auxiliary examinations, a final auxiliary decision is made.

[0024] Optionally, in S1, the disease is characterized by cough symptoms in children.

[0025] Optionally, in S2, the etiology diagnosis question group includes: cough variant asthma diagnosis questions, allergic cough diagnosis questions, asthma diagnosis questions, upper airway cough syndrome diagnosis questions, gastroesophageal reflux-related cough diagnosis questions, post-infectious cough diagnosis questions, psychogenic / tic-related cough diagnosis questions, pertussis diagnosis questions, pulmonary tuberculosis diagnosis questions, persistent bacterial bronchitis diagnosis questions, bronchiectasis diagnosis questions, persistent and chronic pneumonia diagnosis questions, cystic fibrosis diagnosis questions, congenital laryngotracheomalacia diagnosis questions, primary ciliary dyskinesia diagnosis questions, and drug-induced cough diagnosis questions.

[0026] Optionally, see Table 1. The feature mapping rule in S3 is the 6T-6S feature mapping rule; the 6T-6S features include: 6T entry features and 6S entry features. The 6T features include onset time, triggering and aggravating factors, disease course, cough type, accompanying symptoms, and previous treatment response; the 6S features include sneezing and allergies, swallowing and bowel movements, sleep, movement, severe-persistent-rare-recurrent (SPUR) and voice characteristics.

[0027] Table 1. 6T–6S Item Feature Definition Table

[0028] Referring to Table 2, the 6T items are six categories of information surrounding the cough itself, including: onset time; triggering and aggravating factors, such as whether the patient has recently been exposed to cold air, engaged in strenuous exercise, lay down for a long time, or eaten; disease course, such as whether the condition is recurrent or prolonged; cough type, such as dry cough or wet cough; accompanying symptoms, including fever, wheezing, nasal congestion and runny nose, acid reflux and vomiting; and previous treatment response, such as whether antibiotics or inhaled hormones have been used, and whether the response was effective.

[0029] The 6S criteria are six categories of clues that quickly indicate different causes of illness, including sneezing and allergies (suggesting allergy-related conditions and UACS, etc.); abnormal swallowing and bowel movements (suggesting swallowing problems, foreign body aspiration, or gastrointestinal issues); sleep disturbances or nocturnal exacerbations (often suggesting asthma-related coughs or CVA, etc.); exercise-induced or aggravated symptoms (suggesting airway hyperresponsiveness, possibly asthma); severe, persistent, rare, or recurrent symptoms (suggesting immunodeficiency or requiring further investigation); and distinctive cough sounds (barking, croaking, hoarseness, etc., suggest upper airway disease or specific causes).

[0030] Table 2. Specific mapping rules between 6T-6S and common causes of disease

[0031] Furthermore, the 6T-6S feature mapping rule determination process includes: Etiological clues were extracted based on guidelines / consensus and clinical diagnostic criteria and converted into cough characteristic options; Through discussions among clinical experts, cough feature options were categorized, merged, and redundancy removed to obtain the basic information dimension and the quick prompt clue dimension for cough. The 6T+6S feature mapping rule is determined based on the dimensions of basic cough information and quick cueing.

[0032] Furthermore, the questionnaire in S1 was designed based on cough characteristic options.

[0033] Optionally, the formula for calculating the candidate cause score in S3 is as follows: ; in, j Indicates candidate etiology j , S jk Indicates candidate etiology j The score, I ( option jk ) indicates whether to select.

[0034] During the calculation of candidate etiology scores, I ( option jk () is a binary indicator function used to indicate whether parents have selected a candidate cause of illness. j The relevant first k Each questionnaire feature option. When an option is selected and it meets the etiology scoring rules, I ( option jk If the value is 1; or if the option is not selected, another mutually exclusive option is selected, the option is "none / unclear / not seen", or the question is not triggered due to a logical jump, I ( option jk )=0.

[0035] In the single-choice questions, only the selected option is assigned a value of 1, while the other mutually exclusive options are assigned a value of 0. In the multiple-choice questions, each selected option is assigned a value of 1, and the unselected options are assigned a value of 0. Therefore, the candidate causal score can be represented as the sum of the scores of all relevant selected features of that causal factor, thus achieving a structured and quantifiable analysis of the parents' questionnaire information. I ( option jk The calculation formula for ) is as follows: .

[0036] The answer options for the etiology discrimination question set (i.e., the etiology diagnosis question set in S2) in this invention are derived from the typical clinical manifestations, precipitating factors, accompanying symptoms, and treatment responses summarized in domestic and international guidelines, consensus opinions, and diagnostic criteria for chronic cough in children and related etiologies (such as CVA, asthma, UACS, PBB, PIC, GERC, pertussis, tuberculosis, foreign body aspiration, etc.). Referring to Tables 1 and 2, this invention structurally breaks down the above evidence-based points according to the 6T–6S framework, forming a set of questionnaire options that parents can understand and fill out.

[0037] Based on this, option scores S jk The score is determined by the differential value and strength of evidence of the feature in the etiology: features with high specificity or that can serve as key clues or diagnostic clues are assigned higher scores, while features that are generally supportive are assigned medium scores.

[0038] To reduce subjective weighting bias and improve consistency, this invention introduces a multi-model consistency verification process (DeepSeek, Claude3, GPT-4, and Win-GPT) during the development phase, under the constraints of fixed prompt word templates and unified output format. This process provides structured scoring suggestions for feature-causal association strength / evidence stratification and generates suggested values ​​through consistency rules (averaging), which serve as the final option scores (i.e., the scores for candidate causal factors).

[0039] In conjunction with the above embodiments, in S4 of the present invention, the scores of each possible candidate cause are judged according to a preset threshold to obtain the judgment results corresponding to different candidate causes, i.e., the preliminary cause judgment results. To ensure the explicit feasibility of the threshold determination and multi-cause output steps in S4 of the present invention, a threshold is preset for each candidate cause. T j (Total score for the corresponding question set) Max j ).

[0040] Taking childhood cough as an example, after parents complete the questionnaire, the system calculates scores for each cause. Score ( j ) 。 when Score ( j )≥ T j When the threshold is met, the corresponding cause is output; if multiple causes simultaneously meet the threshold, they are output together. The thresholds / total scores for each cause in the example are as follows (…). T j / Max jCough variant asthma: 5 / 7; Allergic cough: 4 / 8; Asthma: 5 / 10; Upper airway cough syndrome: 4 / 6; Gastroesophageal reflux-related cough: 4 / 6; Post-infectious cough: 5 / 8; Psychogenic / tic-related cough: 5 / 7; Pertussis: 4 / 10; Pulmonary tuberculosis: 4 / 10; Persistent bacterial bronchitis: 4 / 6; Bronchiectasis: 4 / 6; Persistent and chronic pneumonia: 3 / 5; Cystic fibrosis: 4 / 6; Congenital laryngotracheomalacia: 5 / 9; Primary ciliary dyskinesia: 5 / 8; Drug-induced cough: 4 / 6.

[0041] Optionally, the S6 scoring step involves obtaining and recording the confidence score and the help score of the final auxiliary decision; the confidence score is used to characterize the doctor's degree of acceptance of the final auxiliary decision, and the help score is used to characterize the reliability of the final auxiliary decision.

[0042] Optional, S7 validation step: Follow up and evaluate patients based on SOP procedures to obtain optimized rule thresholds and weight annotation data.

[0043] Reference Figure 2 and Figure 3 This paper mainly demonstrates the main process of the embodiments of the present invention in the clinical application and follow-up verification stage, namely, the process of patient enrollment, grouping, questionnaire assistance, doctor judgment, follow-up and research outcome evaluation. Taking the process of providing auxiliary decision-making for the diagnosis of cough in children as an example, the present invention further illustrates a disclosed etiology screening and auxiliary decision-making method based on structured electronic questionnaires.

[0044] Combination Figure 2 This study aims to recruit 400 minors with different cough symptoms. Specifically, participants aged 3–18 years who have presented with a cough lasting ≥4 weeks and have not had a fever in the past 2 weeks will be recruited. After recruitment, accompanying parents will complete the questionnaire. Eligible participants will be excluded, such as those with congenital airway malformations, chronic lung disease, severe immunodeficiency, those recently involved in other interventional clinical trials, and those unable to complete the follow-up as required.

[0045] During the questionnaire completion process, doctors were divided into a routine consultation group and a questionnaire assistance group based on their awareness of the current clinical trial. Doctors in the routine consultation group conducted normal consultations with patients and received a "doctor's diagnosis"; simultaneously, parents completed a questionnaire during the diagnosis process and received "questionnaire etiology hints" for later use. Doctors in the questionnaire assistance group, based on the "questionnaire etiology hints," derived a corresponding "doctor's diagnosis" and, after completing the diagnosis, evaluated the questionnaire's helpfulness in diagnosis or decision-making and the reliability of its auxiliary function. The evaluation range was 1-10 points, with higher scores indicating higher doctor acceptance of the questionnaire.

[0046] The questionnaire includes the following specific information: basic information about the child; duration, nature, and timing of the cough; triggers and aggravating factors of the cough (cold air, exercise, after eating, etc.); accompanying symptoms of the cough (nasal symptoms, wheezing, acid reflux, choking, difficulty swallowing, constipation, and halitosis, etc.); characteristics of the cough sound; past medical history and family history; and responses to previous examinations and treatments.

[0047] During the questionnaire completion process, this invention employs logical branching, triggering different question groups based on the different symptoms initially reported by the children, such as the duration of cough, whether it is a wet cough or purulent sputum, whether it worsens at night, and whether it is accompanied by nasal symptoms. In the subsequent refreshed questions that need to be filled in, irrelevant questions are reduced, and the completion efficiency is improved.

[0048] After the questionnaire is completed, the etiology is scored and output based on the completed questionnaire. In this invention, etiology scoring is based on a rule base, which is a database of mappings and weights established based on the relationship between "key features and etiological direction" by referring to domestic and international guidelines and consensus on chronic cough in children and expert experience.

[0049] Based on the database and the information provided by the etiology thresholds / total scores disclosed in the above embodiments, parallel calculations and scoring of different etiologies are performed, and all possible etiology directions and suggestions are output (e.g., nocturnal or exercise-induced cough suggests cough variant asthma; obvious nasal symptoms suggest upper airway cough syndrome; wet cough, purulent sputum and antibacterial test correlation suggest persistent bacterial bronchitis; post-infectious onset suggests post-infectious cough, etc.).

[0050] After the routine diagnosis and treatment group and the questionnaire-assisted group completed their diagnoses, they shared and discussed their respective diagnostic results, uploaded them to the work group, or used other methods to conduct research and discussion based on the children's medical records.

[0051] After obtaining the preliminary diagnosis of the etiology, the clinical process and follow-up loop are implemented. The complete process of this invention in clinical application is described in detail below: (1) Reference Figure 2 Following a parallel grouping process, in the routine treatment group, doctors conducted routine consultations and made diagnoses without knowing the questionnaire results. Parents completed the questionnaire after the consultation, which generated a questionnaire-based causal suggestion and was used only for research records. In the questionnaire assistance group, parents completed the questionnaire before the consultation, which generated a questionnaire-based causal suggestion and provided it to the doctors for reference. The doctors combined the routine consultation with the questionnaire-based causal suggestion to make a diagnosis and recorded the questionnaire tool's helpfulness in the diagnostic decision-making and the degree of trust in the questionnaire-based causal suggestion.

[0052] (2) Reference Figure 3Two weeks after the initial consultation, researchers followed up with parents via phone or WeChat, recording cough progression, medication use, and treatment response according to a standardized operating procedure (SOP). The accuracy of the doctor's initial diagnosis was assessed based on the follow-up results. Simultaneously, parents rated the consistency between the questionnaire's suggested cause and the doctor's diagnosis, and also rated the effectiveness of the current treatment. If the treatment effectiveness score was insufficient, parents were guided to proceed with further treatment, and a follow-up was conducted again one week later.

[0053] (3) Output indicators include: comparing the difference in diagnostic accuracy between doctors in the routine treatment group and the questionnaire-assisted group; comparing the scores of parents in the two groups on the consistency between the questionnaire's etiology suggestions and doctors' diagnostic results; and analyzing the help and confidence scores of doctors when using the questionnaire to evaluate the clinical feasibility and application value of the tool.

[0054] In the clinical process, the content validity of the questionnaire items also needs to be verified. The specific verification process is as follows: Clinical experts evaluated the relevance of entries in the etiology identification problem; "highly relevant" was scored as 1 point, and "irrelevant" was scored as 0 points. The CVI of each entry and the overall S-CVI / Ave were calculated to screen and optimize entries, and to serve as one of the bases for rule base updates and threshold adjustments.

[0055] Furthermore, the content validity verification of questionnaire items is an optional implementation method in the questionnaire development, item selection and rule base optimization stages of this invention. Its purpose is to explain the basis for the formation and optimization method of questionnaire items and rule base. It is not a necessary step in the main process of clinical application, so it will not be drawn separately, but only described in words.

[0056] The present invention also discloses a disease etiology screening and decision support system based on structured electronic questionnaires, which is used to implement a disease etiology screening and decision support method based on structured electronic questionnaires as described in any of the above embodiments, including a patient-side questionnaire module, a logic branch and feature extraction module, a disease etiology inference and scoring engine, a doctor-side display and decision support module, a follow-up and verification module, and a data storage and security module connected in sequence. The patient-side questionnaire module allows patients or their families to complete questionnaires on mobile devices via QR codes or links, thereby obtaining information about the patient's disease characteristics. The logic branching and feature extraction module is used to identify the cause of disease based on disease characteristics, trigger branch questions, and enter the corresponding etiology diagnosis question group. The etiology inference and scoring engine is used to calculate etiology scores for etiology diagnosis questions to obtain a set of candidate etiology scores. Based on etiology thresholds, the engine judges the candidate etiology scores and outputs the etiology that meets the requirements, thus obtaining a preliminary etiology judgment result. The doctor's side display and decision support module is used to present the questionnaire answers and preliminary diagnosis results, and record the doctor's diagnosis results, whether the questionnaire was referenced, the helpness score and the confidence score. The follow-up and verification module is used to conduct follow-up assessments according to the SOP procedure based on the preset follow-up cycle. The data storage and security module is used for data anonymization and encrypted storage, and provides access control and log traceability.

[0057] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for etiological screening and decision support based on structured electronic questionnaires, characterized in that, include: S1 data collection steps: Obtain the patient's disease characteristics through the completed questionnaire; S2 identification steps: Based on the questionnaire jump rules, the etiology is identified using disease characteristics, and the corresponding etiology diagnosis question group is entered; the etiology diagnosis question group includes various etiology diagnosis questions corresponding to the disease characteristics; S3 calculation steps: Based on the feature mapping rules, calculate the etiology score for the etiology diagnosis question set to obtain the candidate etiology score set; The candidate etiology score set includes etiology scores that correspond one-to-one with the etiology; S4 Judgment Step: Based on the etiology threshold, the candidate etiology score set is judged, and the etiology that meets the requirements is output to obtain the preliminary etiology judgment result; S5 auxiliary step: Based on the preliminary judgment of the etiology, combined with the patient's physical examination and auxiliary examinations, a final auxiliary decision is made.

2. The method for etiological screening and decision support based on a structured electronic questionnaire according to claim 1, characterized in that, In S1, the disease is characterized by cough symptoms in children.

3. The method for etiological screening and decision support based on a structured electronic questionnaire according to claim 2, characterized in that, In S2, the etiology diagnosis question group includes: cough variant asthma diagnosis, allergic cough diagnosis, asthma diagnosis, upper airway cough syndrome diagnosis, gastroesophageal reflux-related cough diagnosis, post-infectious cough diagnosis, psychogenic / tic-related cough diagnosis, pertussis diagnosis, pulmonary tuberculosis diagnosis, persistent bacterial bronchitis diagnosis, bronchiectasis diagnosis, persistent and chronic pneumonia diagnosis, cystic fibrosis diagnosis, congenital laryngotracheomalacia diagnosis, primary ciliary dyskinesia diagnosis, and drug-induced cough diagnosis.

4. The method for etiological screening and decision support based on a structured electronic questionnaire according to claim 2, characterized in that, The feature mapping rule in S3 is the 6T-6S feature mapping rule; the 6T-6S features include: 6T entry features and 6S entry features; The 6T features include onset time, triggering and aggravating factors, disease course, cough type, accompanying symptoms, and previous treatment response; the 6S features include sneezing and allergies, swallowing and bowel movements, sleep, movement, severity-persistence-rare-recurrence, and voice characteristics.

5. The method for etiological screening and decision support based on a structured electronic questionnaire according to claim 4, characterized in that, The 6T-6S feature mapping rule determination process includes: Etiological clues were extracted based on guidelines / consensus and clinical diagnostic criteria and converted into cough characteristic options; Through discussions among clinical experts, cough feature options were categorized, merged, and redundancy removed to obtain the basic information dimension and the quick prompt clue dimension for cough. The 6T+6S feature mapping rule is determined based on the dimensions of basic cough information and quick cueing.

6. The method for etiological screening and decision support based on a structured electronic questionnaire according to claim 5, characterized in that, The questionnaire in S1 was designed based on cough characteristic options.

7. The method for etiological screening and decision support based on a structured electronic questionnaire according to claim 2, characterized in that, The formula for calculating the candidate etiology score in S3 is as follows: ； in, j Indicates candidate etiology j , S jk Indicates candidate etiology j The score, I ( option jk ) indicates whether to select.

8. The method for etiological screening and decision support based on a structured electronic questionnaire according to claim 1, characterized in that, Also includes: S6 scoring steps: Obtain and record the confidence score and the help score for the final decision support; The confidence score is used to characterize the degree to which doctors approve of the final auxiliary decision, while the help score is used to characterize the reliability of the final auxiliary decision.

9. A method for etiological screening and decision support based on a structured electronic questionnaire according to claim 1, characterized in that, Also includes: S7 Validation Steps: Follow up and evaluate patients based on SOP procedures to obtain optimized rule thresholds and weight annotation data.

10. A system for etiological screening and decision support based on structured electronic questionnaires, characterized in that, The method for implementing a structured electronic questionnaire-based etiology screening and decision support method as described in any one of claims 1-9 includes a patient-side questionnaire module, a logic branching and feature extraction module, an etiology inference and scoring engine, a doctor-side display and decision support module, a follow-up and verification module, and a data storage and security module connected in sequence. The patient-side questionnaire module allows patients or their families to complete questionnaires on mobile devices via QR codes or links, thereby obtaining information about the patient's disease characteristics. The logic branching and feature extraction module is used to identify the cause of disease based on disease characteristics, trigger branch questions, and enter the corresponding etiology diagnosis question group. The etiology inference and scoring engine is used to calculate etiology scores for etiology diagnosis questions to obtain a set of candidate etiology scores. Based on etiology thresholds, the engine judges the candidate etiology scores and outputs the etiology that meets the requirements, thus obtaining a preliminary etiology judgment result. The doctor's side display and decision support module is used to present the questionnaire answers and preliminary diagnosis results, and record the doctor's diagnosis results, whether the questionnaire was referenced, the helpness score and the confidence score. The follow-up and verification module is used to conduct follow-up assessments according to the SOP procedure based on the preset follow-up cycle. The data storage and security module is used for data anonymization and encrypted storage, and provides access control and log traceability.

Images