Intelligent processing method and system for urinary surgery diagnosis and treatment data

Abstract

The invention belongs to the technical field of medical data processing, and relates to a urinary surgery diagnosis and treatment data intelligent processing method and system, and the system comprises a multi-modal data collection module which is used for collecting basic diagnosis and treatment data, behavior data and diagnosis and treatment process data of a patient; the diagnosis and treatment data intelligent analysis module is used for calculating the intention value of the patient for the diagnosis and treatment item and generating a diagnosis and treatment path diagram; the intelligent decision support module is used for generating personalized diagnosis and treatment suggestions in combination with a guide and a model; the doctor-patient interaction feedback module is used for collecting patient feedback and updating an analysis weight; and the follow-up management module is used for generating a follow-up plan and relapse early warning. Basic diagnosis and treatment data and patient behavior data are fused, and diagnosis and treatment requirements are reflected more comprehensively; the dimension weight is adjusted and analyzed in real time through patient feedback, and the personalized diagnosis and treatment precision is improved; a guide rule and a machine learning model are combined to assist a doctor to quickly generate an optimal treatment scheme; through the block chain and privacy protection technology, compliance use of the medical data is ensured.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical data processing, and in particular, relates to an intelligent processing method and system for urology diagnosis and treatment data. Background Art

[0002] In current urology clinical practice, the processing and application of diagnosis and treatment data face multi-dimensional challenges, which are specifically reflected in the following aspects:

[0003] The data collection dimension is single, covering only basic medical records and examination reports, and lacks in-depth capture of patient behavior data (such as the duration of symptom self-examination and the attention paid to examination items);

[0004] Data analysis remains at the statistical level and cannot dynamically generate personalized treatment recommendations based on diagnosis and treatment guidelines and clinical pathways;

[0005] Patient follow-up relies on manual operations and lacks proactive intervention mechanisms based on data predictions;

[0006] The lack of a doctor-patient interaction module makes it difficult to obtain real-time patient feedback on treatment plans and optimize the treatment path.

[0007] To this end, the present invention provides a method and system for intelligent processing of urology diagnosis and treatment data. Summary of the Invention

[0008] The purpose of the present invention is to provide an intelligent processing method and system for urology diagnosis and treatment data, which solves the problems existing in the prior art.

[0009] The purpose of the present invention can be achieved through the following technical solutions:

[0010] An intelligent processing system for urology diagnosis and treatment data, comprising:

[0011] Multimodal data collection module, used to collect patients' basic diagnosis and treatment data, behavioral data, and diagnosis and treatment process data;

[0012] Intelligent analysis module for diagnosis and treatment data, used to calculate the patient's intention value for diagnosis and treatment items and generate a diagnosis and treatment pathway map;

[0013] An intelligent decision support module, used to combine guidelines and models to generate personalized diagnosis and treatment recommendations; a doctor-patient interactive feedback module, used to collect patient feedback and update analysis weights;

[0014] Follow-up management module, used to generate follow-up plans and recurrence warnings.

[0015] Preferably, the multimodal data acquisition module includes:

[0016] Medical record document collection unit, used to obtain structured medical record data such as chief complaint and current medical history;

[0017] The behavioral data collection unit records the duration of symptom self-examination, the duration of browsing examination items, and the number of treatment plan searches through the patient-side APP;

[0018] System docking unit, used to synchronize diagnosis and treatment process data in the hospital's HIS / LIS / PACS.

[0019] Preferably, the diagnosis and treatment data intelligent analysis module includes:

[0020] Weight setting unit, used to initialize the weight coefficients of basic data, behavior data, and search data;

[0021] Intention Assessment Unit, based on formula Calculate the intention to treat value;

[0022] The path generation unit generates a line graph and threshold marks with the diagnosis and treatment items as the X-axis and the intention value as the Y-axis.

[0023] Preferably, the intelligent decision support module includes:

[0024] A rule engine unit, used to generate preliminary recommendations based on urology diagnosis and treatment guidelines;

[0025] Machine learning unit, which trains diagnosis and treatment plan prediction models based on historical cases;

[0026] The report generation unit outputs a visual decision-making report including disease assessment and plan comparison.

[0027] Preferably, the doctor-patient interaction feedback module includes:

[0028] The questionnaire design unit includes three scoring questions: symptom description accuracy, examination item matching, and treatment plan acceptance;

[0029] Weight update unit, dynamically adjusts p1, p2, p3 based on feedback scores;

[0030] Preference management unit, which records patient feedback history to optimize recommendation strategies.

[0031] Preferably, the follow-up management module includes:

[0032] Risk prediction unit, using the Cox model to predict recurrence risk;

[0033] Plan generation unit, which develops personalized follow-up plans based on risk levels;

[0034] The early warning trigger unit prompts doctors to intervene when follow-up data is abnormal.

[0035] Preferably, it also includes a data security and privacy protection module, which uses blockchain encryption storage, hierarchical access control and de-identification technology to ensure data security.

[0036] The present invention also discloses an intelligent processing method for urology diagnosis and treatment data, comprising:

[0037] Data collection steps to obtain multi-dimensional diagnosis and treatment data;

[0038] Intelligent analysis steps to calculate intention value and generate diagnosis and treatment pathways;

[0039] Decision support steps, combining guidelines with model output recommendations;

[0040] interactive feedback step, updating weights based on patient scores;

[0041] Follow up management steps, implement risk prediction and dynamic monitoring.

[0042] Preferably, in the intelligent analysis step, a browsing stay threshold Δt is set, and when the browsing time s of the inspection item is n When it is greater than Δt, it is considered as a valid stay and included in the total stay time t 停留 .

[0043] Preferably, in the interactive feedback step, the formula Update the basic data weight coefficient, where Q1 is the symptom description satisfaction score.

[0044] The beneficial effects of the present invention are as follows: The present invention integrates basic diagnosis and treatment data with patient behavior data to more comprehensively reflect diagnosis and treatment needs; adjusts the analysis dimension weights in real time through patient feedback to improve the accuracy of personalized diagnosis and treatment; combines guideline rules with machine learning models to assist doctors in quickly generating optimal treatment plans; and ensures the compliance of medical data use through blockchain and privacy protection technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0046] Figure 1 This is a system block diagram of an intelligent processing system for urology diagnosis and treatment data of the present invention;

[0047] Figure 2 The present invention provides a flow chart of an intelligent processing method for urology diagnosis and treatment data. DETAILED DESCRIPTION

[0048] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative efforts shall fall within the scope of protection of the present invention.

[0049] See also Figure 1 As shown, the present invention is an intelligent processing system for urology diagnosis and treatment data, comprising:

[0050] A multimodal data collection module is used to collect basic patient diagnosis and treatment data, behavioral data, and diagnosis and treatment process data. The multimodal data collection module includes: a medical record collection unit for obtaining structured medical record data such as the chief complaint and current medical history; a behavioral data collection unit for recording the duration of symptom self-examination, the duration of browsing examination items, and the number of treatment plan searches through the patient-side APP; and a system docking unit for synchronizing diagnosis and treatment process data in the hospital's HIS / LIS / PACS.

[0051] The diagnosis and treatment data intelligent analysis module is used to calculate the patient's intention value for the diagnosis and treatment items and generate a diagnosis and treatment path map. The diagnosis and treatment data intelligent analysis module includes: a weight setting unit for initializing the weight coefficients of basic data, behavior data, and search data; an intention evaluation unit based on the formula Calculate the diagnosis and treatment intention value; the path generation unit generates a line graph and threshold marks with the diagnosis and treatment items as the X-axis and the intention value as the Y-axis.

[0052] The intelligent decision support module is used to combine guidelines and models to generate personalized diagnosis and treatment recommendations; the doctor-patient interactive feedback module is used to collect patient feedback and update analysis weights. The intelligent decision support module includes: a rule engine unit for matching urology diagnosis and treatment guidelines to generate preliminary recommendations; a machine learning unit for training diagnosis and treatment plan prediction models based on historical cases; a report generation unit for outputting a visual decision report containing condition assessment and plan comparison. The doctor-patient interactive feedback module includes: a questionnaire design unit, which includes three scoring questions: symptom description accuracy, examination item matching, and treatment plan acceptance; a weight update unit, which dynamically adjusts p1, p2, and p3 based on feedback scores; and a preference management unit, which records patient feedback history to optimize recommendation strategies.

[0053] The follow-up management module is used to generate follow-up plans and recurrence warnings. The follow-up management module includes a risk prediction unit that uses the Cox model to predict recurrence risk; a plan generation unit that develops personalized follow-up plans based on risk levels; and an early warning trigger unit that prompts doctors to intervene when follow-up data is abnormal.

[0054] It also includes a data security and privacy protection module, which uses blockchain encryption storage, hierarchical access control and de-identification technology to ensure data security.

[0055] See also Figure 2 As shown, the present invention also discloses an intelligent processing method for urology diagnosis and treatment data, comprising:

[0056] Data collection steps to obtain multi-dimensional diagnosis and treatment data;

[0057] Intelligent analysis step, calculate the intention value and generate the diagnosis and treatment path; in the intelligent analysis step, set the browsing stay threshold Δt, when the browsing time of the inspection item s n When it is greater than Δt, it is considered as a valid stay and included in the total stay time t 停留 .

[0058] Decision support steps, combining guidelines with model output recommendations;

[0059] The interactive feedback step updates the weight based on the patient score; in the interactive feedback step, the formula Update the basic data weight coefficient, where Q1 is the symptom description satisfaction score.

[0060] Follow up management steps, implement risk prediction and dynamic monitoring.

[0061] The specific embodiments are as follows:

[0062] Example 1

[0063] Multimodal data acquisition module

[0064] Medical record collection unit:

[0065] In addition to the chief complaint and current medical history, structured medical record data such as past medical history, allergy history, family history, physical examination results, laboratory test reports (such as urine routine, renal function), and imaging examination data (such as ultrasound and CT images) were further collected.

[0066] Use natural language processing (NLP) technology to parse unstructured text (such as handwritten medical records by doctors) and convert it into standardized data fields.

[0067] Behavioral data collection unit:

[0068] The patient-side APP has added behavioral indicators such as filling records of symptom self-assessment scales (such as IPSS scoring scales), number of times treatment plans are collected / shared, and length of online communication with medical staff.

[0069] Use tracking technology to monitor the user's operation trajectory in the APP, such as the number of clicks on specific diagnosis and treatment items, the sliding depth, and the frequency of repeated visits.

[0070] System docking unit:

[0071] Real-time data synchronization with the hospital's HIS / LIS / PACS system, including registration information, examination appointment records, prescription circulation status, fee settlement data and other information on the entire diagnosis and treatment process.

[0072] It supports connecting to third-party health management platforms (such as smart wearable device data) through API interfaces to obtain patients' daily physiological indicators (such as blood pressure and heart rate).

[0073] Intelligent analysis module for diagnosis and treatment data

[0074] Weight setting unit:

[0075] When initializing weight coefficients, we incorporate expert experience and historical data statistics. For example, the default weight for basic data (medical records) is p1 = 0.5, for behavioral data (app operations) p2 = 0.3, and for search data (treatment plan queries) p3 = 0.2. These can be adjusted dynamically based on department needs.

[0076] Intention Assessment Unit:

[0077] Parameter definition expansion in formula:

[0078] t n : The completeness score of the medical record data for the nth diagnosis and treatment item (e.g., 0-10 points, calculated based on the completion of required fields).

[0079] s n : The effective browsing time of the nth inspection item (must exceed the preset threshold Δt = 30 seconds to avoid accidental touch interference).

[0080] k n : The number of searches for the nth treatment option, combined with a time decay factor (e.g., searches in the last 7 days have a higher weight).

[0081] Path generation unit:

[0082] When generating a diagnosis and treatment pathway diagram, the recommended pathway in the guidelines (such as the standard process of the urinary stone diagnosis and treatment guidelines) is superimposed and compared with the patient's personalized pathway, and key nodes (such as examinations, surgery, and medication) are marked with different colors.

[0083] Threshold marking uses dynamic warning lines (such as intention values ​​> 0.7 for high-priority projects and < 0.3 for excluded projects).

[0084] Intelligent decision support module

[0085] Rule Engine Unit:

[0086] It has a built-in knowledge base of urology diagnosis and treatment guidelines (such as the "Guidelines for the Diagnosis and Treatment of Urology and Andrology Diseases in China"), which supports keyword search and condition matching (such as patient age, stone size, and renal function status).

[0087] When generating preliminary recommendations, drugs that conflict with the patient's allergy history and treatment plans that conflict with the underlying disease are automatically excluded.

[0088] Machine Learning Unit:

[0089] The random forest algorithm is used to train the diagnosis and treatment plan prediction model, and the input features include: medical record data, behavioral data, laboratory indicators, imaging features, etc.

[0090] Regularly update the model with new case data to improve the prediction accuracy of rare diseases such as adrenal tumors.

[0091] Report generation unit:

[0092] The visual decision report includes charts such as Sankey diagrams (showing the flow of diagnosis and treatment pathways), radar charts (comparing the advantages and disadvantages of plans), and survival curve charts (predicting the probability of postoperative recurrence).

[0093] Provides doctor annotation function and supports adding personalized adjustment suggestions in the report.

[0094] Doctor-patient interaction feedback module

[0095] Questionnaire Design Unit:

[0096] The rating questions were expanded to a 5-point Likert scale (1 = very dissatisfied, 5 = very satisfied), with new questions added, such as “timeliness of follow-up services” and “satisfaction with privacy protection”.

[0097] Supports voice input to text to fill out questionnaires, improving the convenience for elderly patients.

[0098] Weight update unit:

[0099] The dynamic adjustment formula is expanded to Among them, w1, w2, and w3 are the importance weights of each scoring item (such as symptom description accuracy w1 = 0.4, treatment plan acceptance w3 = 0.3).

[0100] Preferences snap-in:

[0101] Establish a patient preference tag library (such as "prioritize non-invasive treatment" and "refuse hormonal drugs") to automatically filter out options that do not meet preferences when recommending subsequent plans.

[0102] Follow-up management module

[0103] Risk Prediction Unit:

[0104] The input parameters of the Cox model include: pathological type, surgical method, postoperative PSA value (for prostate cancer patients), number of comorbidities, etc., and the output is the recurrence risk probability value and confidence interval.

[0105] Plan generation unit:

[0106] Develop a follow-up plan based on risk level:

[0107] Low risk: Urinalysis + ultrasound every 6 months;

[0108] Medium risk: specialist examination + tumor marker testing every 3 months;

[0109] High risk: Monthly imaging review + doctor’s consultation.

[0110] Early warning trigger unit:

[0111] Definition of abnormal data: If the PSA value increases by >2 ng / mL twice in a row or a new mass is found on ultrasound, a three-level warning will be triggered (yellow reminds the nurse to follow up, red prompts the attending physician to intervene).

[0112] Data security and privacy protection module

[0113] Application of blockchain technology: Adopting alliance chain architecture, recording data access logs (including operator, time, and data range) to ensure that operations are traceable.

[0114] Hierarchical access control: Set data viewing permissions based on the roles of medical staff (such as interns, attending physicians, and department heads). Sensitive information (such as patient names and ID numbers) is hidden by default and requires secondary authentication to unlock.

[0115] De-identification technology: Patient data is irreversibly encrypted (such as the SHA-256 hash algorithm) to retain only the minimal identification required for the research (such as an anonymous ID + a fragment of the date of birth).

[0116] Example 2

[0117] Data collection steps

[0118] New multi-source data verification mechanism: Compare the consistency of medical record data and APP behavior data (for example, if the symptom of "frequent urination" is recorded in the medical record, if the relevant self-examination record is not triggered in the APP, it will be marked as data abnormality and require manual review).

[0119] Supports offline data import (such as taking a photo of an external hospital inspection report and uploading it, and converting it into structured data through OCR technology).

[0120] Intelligent analysis steps

[0121] Introduce outlier detection algorithms (such as the IQR method) to filter out noise in behavioral data (for example, a single browsing time > 30 minutes is considered an invalid operation).

[0122] After the diagnosis and treatment pathway is generated, the differences between the order recommended by the guidelines and the order actually intended by the patient are automatically marked, prompting doctors to pay attention to potential communication needs.

[0123] Decision support steps

[0124] When generating personalized recommendations, priority is given to examinations / treatment items covered by medical insurance, and the self-pay ratio is marked for patients' reference.

[0125] Provide cost-effectiveness comparisons of multiple options (e.g., comparison of total costs, length of hospital stay, and recurrence rates between surgical treatment and drug treatment).

[0126] Interactive feedback steps

[0127] Patient feedback and weight updates are linked in real time: for example, if most patients score "examination item matching" below 3 points, the system will automatically increase the weight p2 of "examination item browsing time" in the behavioral data.

[0128] A new doctor-side feedback portal has been added, allowing doctors to score the rationality of treatment pathways for optimizing machine learning models.

[0129] Follow-up management steps

[0130] The follow-up plan reaches patients through multiple channels such as SMS + APP push + phone calls, and supports customized reminder times (such as notification 3 days in advance).

[0131] After the recurrence warning is triggered, a consultation application form (including patient medical history, examination results, and risk assessment report) is automatically generated and pushed to the expert database of the superior hospital.

[0132] Throughout the specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0133] The above content is merely an example and explanation of the concept of the present invention. Those skilled in the art may make various modifications or additions to the described specific embodiments or replace them in a similar manner. As long as they do not deviate from the concept of the invention or exceed the scope defined by the claims, they should all fall within the scope of protection of the present invention.

Claims

1. An intelligent processing system for urology diagnosis and treatment data, characterized by: include: Multimodal data collection module, used to collect patients' basic diagnosis and treatment data, behavioral data, and diagnosis and treatment process data; Intelligent analysis module for diagnosis and treatment data, used to calculate the patient's intention value for diagnosis and treatment items and generate a diagnosis and treatment pathway map; Intelligent decision support module, used to combine guidelines and models to generate personalized diagnosis and treatment recommendations; Doctor-patient interaction feedback module, used to collect patient feedback and update analysis weights; Follow-up management module, used to generate follow-up plans and recurrence warnings.

2. The intelligent processing system for urology diagnosis and treatment data according to claim 1, characterized in that: The multimodal data acquisition module includes: Medical record document collection unit, used to obtain structured medical record data such as chief complaint and current medical history; The behavioral data collection unit records the duration of symptom self-examination, the duration of browsing examination items, and the number of treatment plan searches through the patient-side APP; System docking unit, used to synchronize diagnosis and treatment process data in the hospital's HIS / LIS / PACS.

3. The intelligent processing system for urology diagnosis and treatment data according to claim 1, characterized in that: The diagnosis and treatment data intelligent analysis module includes: Weight setting unit, used to initialize the weight coefficients of basic data, behavior data, and search data; Intention Assessment Unit, based on formula Calculate the intention to treat value; The path generation unit generates a line graph and threshold marks with the diagnosis and treatment items as the X-axis and the intention value as the Y-axis.

4. The intelligent processing system for urology diagnosis and treatment data according to claim 1, characterized in that: The intelligent decision support module includes: A rule engine unit, used to generate preliminary recommendations based on urology diagnosis and treatment guidelines; Machine learning unit, which trains diagnosis and treatment plan prediction models based on historical cases; The report generation unit outputs a visual decision-making report including disease assessment and plan comparison.

5. The intelligent processing system for urology diagnosis and treatment data according to claim 1, characterized in that: The doctor-patient interaction feedback module includes: The questionnaire design unit includes three scoring questions: symptom description accuracy, examination item matching, and treatment plan acceptance; Weight update unit, dynamically adjusts p1, p2, p3 based on feedback scores; Preference management unit, which records patient feedback history to optimize recommendation strategies.

6. The intelligent processing system for urology diagnosis and treatment data according to claim 1, characterized in that: The follow-up management module includes: Risk prediction unit, using the Cox model to predict recurrence risk; Plan generation unit, which develops personalized follow-up plans based on risk levels; The early warning trigger unit prompts doctors to intervene when follow-up data is abnormal.

7. The intelligent processing system for urology diagnosis and treatment data according to claim 1, characterized in that: It also includes a data security and privacy protection module, which uses blockchain encryption storage, hierarchical access control and de-identification technology to ensure data security.

8. A method for intelligently processing urology diagnosis and treatment data, characterized by: An intelligent processing system for urology diagnosis and treatment data as described in any one of claims 1 to 7, comprising: Data collection steps to obtain multi-dimensional diagnosis and treatment data; Intelligent analysis steps to calculate intention value and generate diagnosis and treatment pathways; Decision support steps, combining guidelines with model output recommendations; interactive feedback step, updating weights based on patient scores; Follow up management steps, implement risk prediction and dynamic monitoring.

9. The intelligent processing method for urology diagnosis and treatment data according to claim 8, characterized in that: In the intelligent analysis step, a browsing stay threshold Δt is set. When the browsing time s of the inspection item is n When it is greater than Δt, it is considered as a valid stay and included in the total stay time t 停留 .

10. The intelligent processing method for urology diagnosis and treatment data according to claim 8, characterized in that: In the interactive feedback step, the formula Update the basic data weight coefficient, where Q1 is the symptom description satisfaction score.

Images