A method and system for automatically interpreting and managing tumor gene detection reports

Through the automated interpretation and management system of tumor gene detection reports, the problems of insufficient automation and low data management efficiency in the existing technology are solved, efficient and accurate report interpretation and management are achieved, manual participation is reduced, and the quality of reports is improved. Security.

CN118629575BActive Publication Date: 2024-11-19GENETRON HEALTH (BEIJING) CO LTD +2
View PDF 0 Cites 1 Cited by

Patent Information

Application Number
CN202411087320.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-09
Publication Date
2024-11-19
Estimated Expiration
2044-08-09

AI Technical Summary

Technical Problem

The existing tumor gene detection reporting system is insufficiently automated, incomplete coverage of work scenarios, and limited scope of application, resulting in low quality and efficiency of report interpretation, risk of human error, and low data management and collaboration efficiency, making it difficult to ensure the consistency and data of reports Safety.

Method used

Provide a system for automatic interpretation and management of tumor gene detection reports, including template creation unit, data management unit and report review unit, supporting personalized report template creation, automatic interpretation report generation and online review management, realizing the full process of report automation, Reduce manual participation and improve interpretation efficiency and accuracy.

Benefits of technology

It realizes the rapid, efficient, accurate, comprehensive and standardized interpretation and issuance of tumor gene detection reports, reduces manual errors, improves the consistency and data security of reports, and enhances the degree of automation and scope of application of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118629575B_ABST
    Figure CN118629575B_ABST
Patent Text Reader

Abstract

The present invention relates to a method and system for automatic interpretation and management of tumor gene detection reports, the system comprising: a template creation unit, a data management unit and a report review unit; the template creation unit is used to create a personalized report template according to clinical needs; the data management unit is used to obtain upstream data and medical interpretation information of patient test samples based on the personalized report template, generate an automated interpretation report according to preset data processing and interpretation logic, and store data; the report review unit is used to create a report review task according to the generated automated interpretation report, and send it to a corresponding review client according to preset scheduling management rules, and the corresponding review client reviews it according to preset report review management rules, and manages the entire report review process at the same time, and obtains a final version of the interpretation report of the patient test sample. The present invention can be widely used in the field of medical data processing.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to a method and system for automatic interpretation and management of tumor gene detection reports, and belongs to the field of bioinformatics. Background Art

[0002] With the development and progress of sequencing technology, genetic testing products in the field of precision diagnosis and treatment are constantly emerging at home and abroad, and the requirements for testing items, report analysis content and presentation format, reporting cycle, etc. of tumor genetic testing products are becoming increasingly strict and standardized.

[0003] At present, there are various reporting systems on the market that can assist in the semi-automatic / automatic issuance of genetic testing reports, but the relevant reporting systems still have problems such as insufficient automation, incomplete coverage of work scenarios, and limited scope of application. In the complete process of report production and delivery, many tasks are not supported by the reporting system, but rely on manual participation, such as: basic information retrieval and entry, product template modification and configuration, medical evidence screening to report, reviewer scheduling and task allocation, division of labor and collaboration among multiple reviewers, after-sales report updates and other related work. Manual participation will introduce the risk of human error, which seriously restricts the quality and efficiency of report interpretation and production.

[0004] The current reporting system has the following common problems:

[0005] 1. Lack of automated functions to simultaneously support the hierarchical analysis of tumor gene system variation and / or germline variation, and output comprehensive clinical prompt reports such as targeted therapy, immunotherapy, chemotherapy drugs, clinical trials, genetic prompts, and prognosis by typing;

[0006] 2. Relying on manual differentiation of different test product report templates, retrieving and retrieving medical interpretation information, low operation efficiency and accuracy, slow report generation speed, and repeated confirmation of relevant information in each link of production to ensure the smooth completion of the whole process;

[0007] 3. The incomplete system functions introduce too many manual interventions. However, due to factors such as differences in work experience and qualifications among personnel, there are often subjective differences in the interpretation and issuance of reports by different personnel, which makes it difficult to unify the presentation standards of the medical content of the reports, affecting the consistency of the reports;

[0008] 4. The report production process involves task allocation, collaboration and task management among multiple departments or multiple groups within a department. Communication and collaboration beyond the functions supported by the report system are prone to delays and unreasonable resource allocation, making it difficult to clarify task progress in a timely manner, resulting in low efficiency in each link and affecting report delivery;

[0009] 5. It is difficult to manage massive amounts of medical report data locally offline, which can easily lead to data security issues such as incomplete data, loss, damage, and leakage. It is difficult to trace, review, and reproduce report-related work, making it difficult to meet quality standards. Summary of the invention

[0010] In view of the above problems, the purpose of the present invention is to provide a method and system for automated interpretation and management of tumor gene test reports, which is suitable for the automated and accurate interpretation of reports by genetic testing institutions, has the characteristics of high degree of automation, full coverage of work scenarios, and wide scope of application, and can support the rapid, efficient, accurate, comprehensive, and standardized interpretation and issuance of tumor gene test reports, and can complete one-stop multi-link operations such as automatic interpretation of reports, collaborative review and delivery of reports in the reporting system.

[0011] To achieve the above object, the present invention adopts the following technical solutions:

[0012] In a first aspect, the present invention provides a system for automated interpretation and management of tumor gene test reports, comprising: a template creation unit, a data management unit and a report review unit; the template creation unit is used to create a personalized report template according to clinical needs; the data management unit is used to obtain upstream data and medical interpretation information of patient test samples based on the personalized report template, generate an automated interpretation report according to preset data processing and interpretation logic, and store report-related data; the report review unit is used to create a report review task based on the generated automated interpretation report, and send it to a corresponding review client according to preset scheduling management rules, and the corresponding review client conducts an audit according to the preset report review management rules, and manages the entire process of the report review task to obtain a final interpretation report of the patient test sample.

[0013] Furthermore, the template creation unit includes a module management module and a template management module; the module management module is used to establish and store all sub-modules in the report template, and freely combine different sub-modules to generate a basic report framework corresponding to different products; the template management module is used to create a template name according to clinical needs and associate it with the basic report framework corresponding to different products, and generate or pre-generate a personalized report template.

[0014] Further, the module management module at least includes a targeted medication guidance submodule, a typing / prognosis assessment submodule, a variation analysis submodule, an immunotherapy guidance submodule, a genetic risk prompting submodule, a chemotherapy drug guidance submodule and a clinical trial submodule; the targeted medication guidance submodule is used to guide targeted medication; the typing / prognosis assessment submodule is used to assist in molecular typing and prognosis assessment; the variation analysis submodule is used to analyze the biological functions and carcinogenic effects of gene mutations in various tumor types; the immunotherapy guidance submodule is used to guide immunotherapy; the genetic risk prompting submodule is used to indicate germline variation grading and tumor genetic susceptibility risk; the chemotherapy drug guidance submodule is used to guide the use of chemotherapy drugs; the clinical trial submodule is used to guide the enrollment of tumor patients in clinical trials;

[0015] The template management module includes a template creation submodule, an association information submodule and a report generation submodule; the template creation submodule is used to create a template name and number it according to clinical needs; the association information submodule is used to associate upstream data with the created template; the report generation submodule is used to associate the template name and the report basic framework corresponding to different products created in the module management module to form a personalized report template.

[0016] Furthermore, the data management unit includes a data retrieval module, a data processing module and a data storage module; the data retrieval module is used to obtain upstream data and medical interpretation information of patient test samples according to a personalized report template; the data processing module is used to process the upstream data and medical interpretation information according to preset data processing and interpretation logic to generate an automated interpretation report; the data storage module is used to store report-related data, including the upstream data and medical interpretation information processed by the data processing module.

[0017] Furthermore, the report review unit includes a scheduling management module, a report management module and a task management module; the scheduling management module is used to create a schedule according to preset scheduling management rules, and automatically assign the report review task to the reviewer on duty that day according to preset allocation principles; the report management module is used for the reviewer on duty that day to review the automated interpretation report according to the preset report review management rules and supervise and record the status of the automated interpretation report to obtain the final version of the interpretation report of the patient's test sample; the task management module is used to track the entire process of the report review task in real time.

[0018] Furthermore, the shift management module includes an account and authority setting submodule, a shift schedule creation submodule and an audit task allocation submodule; the account and authority setting submodule is used to open corresponding accounts and allocate different permissions for each auditor according to the auditor's work authority and work content; the shift schedule creation submodule is used to create a shift schedule according to the work authority and work content of different auditors, and determine the auditors on duty on that day; the audit task allocation submodule is used to set audit lines suitable for different scenarios according to different product characteristics, and automatically allocate report audit tasks to auditors on duty on that day according to preset allocation principles;

[0019] The report management module includes an audit submodule and a process management submodule; the audit submodule is used for each auditor to manually review the automated interpretation report according to the preset report audit management rules, and deliver the final version of the interpretation report of the patient test sample after the audit is completed; the process management submodule is used to supervise and record the status of the automated interpretation report, and the record content includes patient test sample information, product name, institution, template name, cancer type, task type, task rollback type, initial review status, review review status, report generation time, report generation status, sending status, bioinformatics review status and audit path;

[0020] The task management module includes an online tracking submodule and a report review and statistics submodule; the online tracking submodule is used to store and view detailed information of each report review task, and to track the entire process of each report review task in real time, including task status, task type, review progress, reviewer, review time, and whether there is an abnormality; the report review and statistics submodule is used to perform multi-dimensional review and statistics on genetic testing reports in the system.

[0021] In a second aspect, the present invention provides a method for automatically interpreting and managing tumor gene detection reports, comprising the following steps:

[0022] 1) Create personalized report templates using the template creation unit based on clinical needs;

[0023] 2) Based on the personalized report template, the data management unit is used to obtain the upstream data and medical interpretation information of the patient's test samples, and an automated interpretation report is generated according to the preset data processing and interpretation logic, and the report data is stored;

[0024] 3) The report review unit creates a report review task based on the automated interpretation report, and sends it to the corresponding review client according to the preset scheduling management rules. The corresponding review client reviews the report according to the preset report review management rules, and manages the entire process of the report review task to obtain the final interpretation report of the patient's test sample.

[0025] Furthermore, in step 1), according to clinical needs, a personalized report template is created using a template creation unit, including the following steps:

[0026] 1.1) Create corresponding sub-modules in the module management module, and freely combine the sub-modules to generate the basic report framework corresponding to different products;

[0027] 1.2) Create a template name in the template management module and associate test information according to clinical needs;

[0028] 1.3) Associate the template name created in the template management module with the report basic framework created in the module management module to generate a personalized report template.

[0029] Furthermore, in step 2), based on the personalized report template, the upstream data and medical interpretation information of the patient's test sample are obtained by using the data management unit, and an automated interpretation report is generated according to the preset data processing and interpretation logic, and the report data is stored, including the following steps:

[0030] 2.1) Based on the basic information and test information of the patient's test samples in the personalized report template, retrieve the pathological diagnosis and test results from the upstream data;

[0031] 2.2) Retrieve the corresponding medical interpretation information based on the pathological diagnosis and test results, and process and interpret the upstream data and medical interpretation information according to the preset data processing and interpretation logic to generate an automated interpretation report.

[0032] Furthermore, in step 2.2), based on the pathological diagnosis and test results, the corresponding medical interpretation information is retrieved, and the upstream data and medical interpretation information are processed and interpreted according to the preset data processing and interpretation logic to generate an automated interpretation report, including:

[0033] 2.2.1) Using pathological diagnosis and test results as input, retrieve corresponding medical interpretation information according to the preset retrieval logic classification;

[0034] Among them, the preset calling logic is:

[0035] ① For target drug matching data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input, and the corresponding targeted drug guidance information is matched in the "target drug matching" category of the medical interpretation information and retrieved into the system, including targeted drugs, drug evidence, evidence type, and sensitivity / resistance;

[0036] ② For classification / prognosis data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input parameters, and the classification / prognosis analysis information is matched in the "classification / prognosis" category of the medical interpretation information and retrieved into the system, including the prognosis assessment results, evidence, and evidence type;

[0037] ③ For mutation analysis data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input parameters, and the mutation analysis description information is matched in the "Mutation Analysis" category of the medical interpretation information and retrieved into the system, including the biological function and carcinogenicity of the mutation;

[0038] ④ For immunotherapy data, the cancer types and gene mutations, TMB, and MSI corresponding to the patient test samples in the upstream data are used as input parameters, and the immunotherapy efficacy-related analytical information is matched in the "immunotherapy" category of the medical interpretation information and retrieved into the system, including whether the gene mutation indicates a positive or negative effect on immunotherapy;

[0039] ⑤ For chemotherapy drug guidance data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input, and the chemotherapy drug guidance information is matched from the "chemotherapy drug" category of the medical interpretation information and retrieved into the system, including the toxic and side effects, effectiveness and medication tips of chemotherapy drugs;

[0040] ⑥ For clinical trial data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input parameters, and the clinical trial data with completely consistent mutations, cancer types and cancer subtypes are matched in the "clinical trial" category of the medical interpretation information and retrieved into the system;

[0041] ⑦ For genetic risk analysis data, the cancer type and test result gene germline variation corresponding to the patient test sample in the upstream data are used as input parameters, and the drug, drug use genetic analysis evidence information, evidence type, and sensitive / resistant information are matched in the "Genetic Analysis" category of the medical interpretation information and retrieved into the system;

[0042] 2.2.2) Process the retrieved medical interpretation information and upstream data according to the preset data processing and interpretation logic, and generate an automated interpretation report according to the personalized report template configuration requirements;

[0043] Among them, the generation of automated interpretation reports includes:

[0044] ① The retrieved data on typing / prognosis, mutation analysis, immunotherapy, chemotherapy drugs, and genetic risk analysis are automatically filled into the corresponding submodules of the personalized report template after processing;

[0045] ② For target drug matching data and clinical trial data, they are processed according to the preset data processing and interpretation logic and then filled into the targeted drug guidance submodule and clinical trial submodule respectively;

[0046] Among them, when processing target drug matching data, it includes:

[0047] Determine the evidence level and variation type of target drug information and related evidence in the medical interpretation information based on preset report display rules and automated judgment logic;

[0048] Screen target drugs and evidence according to preset automated logic rules for screening and sorting target drugs and evidence, as well as the determined evidence level and variation type of each piece of evidence;

[0049] The preset automated logic rules for drug targets and evidence screening and sorting include:

[0050] a. According to the sensitivity or resistance of drugs and evidence, they are automatically displayed in the report in two sections, with all sensitive drugs and evidence in the front and resistant drugs and evidence in the back;

[0051] b. For the same drug, when the medical interpretation information has high-level evidence, the lower-level evidence will not be output;

[0052] c. Evidence for the same drug is presented together;

[0053] d. Sort different evidence by level;

[0054] e. Evidence of the same level is sorted by variant type; clinical trial stages are arranged from high to low;

[0055] f. When encountering contradictory evidence for the same drug, the evidence for target drug analysis shall be ranked according to the level of the evidence itself;

[0056] The screened target drugs and evidence are further screened and sorted according to the preset automated logic for judging, grading and displaying contradictory evidence.

[0057] The present invention adopts the above technical solution, which has the following advantages:

[0058] 1. In the present invention, the data management unit can automatically retrieve, process and store upstream data and medical interpretation information, and display the interpretation content related to the test results of each sample in a standardized form and logic according to established rules, and generate a PDF / Word report with a specific template, which is convenient for the interpreter to review, update, circulate and deliver the automated interpretation report.

[0059] 2. In the present invention, since the template creation unit can freely set sub-modules, the report template framework covers comprehensive scenarios, has diversified report forms, and report types, and can be flexibly configured to meet different scenario requirements, forms, and types.

[0060] 3. The present invention is equipped with a report review unit to conduct online review of the genetic test report and send and deliver the reviewed report, which can efficiently circulate, shorten the time, ensure timeliness, reduce manpower and time costs, optimize resource allocation, and improve the accuracy and timeliness of the test report.

[0061] Therefore, the present invention can be widely applied in the field of medical data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Throughout the accompanying drawings, the same reference numerals are used to represent the same components. In the accompanying drawings:

[0063] Figure 1 This is a system structure diagram for automatic interpretation and management of tumor gene detection reports provided by an embodiment of the present invention;

[0064] Figure 2 It is a structural diagram of a template creation unit in a system for automatic interpretation and management of tumor gene detection reports provided by an embodiment of the present invention;

[0065] Figure 3 It is a structural diagram of a report review unit in a system for automatic interpretation and management of tumor gene detection reports provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0066] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.

[0067] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0068] The interpretation and issuance of tumor gene test reports is a very complex, rigorous, and highly professional job, requiring report interpreters to have professional knowledge backgrounds in biomedicine, tumors, molecular genetics, etc., fully understand and skillfully memorize the interpretation logic of each module of the report, work carefully and meticulously without missing any clinical treatment prompts related to the mutation, and ensure that each report is accurately, comprehensively, and efficiently interpreted. Without the support of high-quality, fully functional automated tools, even professional report interpreters can only manually write and interpret reports issued every day, and are prone to manual writing errors and logical omissions. Therefore, it is extremely important for genetic testing institutions to establish an automated interpretation and reporting system that meets the needs.

[0069] Based on this, in some embodiments of the present invention, a system for automatic interpretation and management of tumor gene detection reports with comprehensive functions, rigorous logic and high automation level (hereinafter referred to as "report system") is provided, which aims to provide functional support in multiple types of application scenarios and improve the quality and efficiency of report interpretation. First, the reporting system is associated with upstream data, and can obtain upstream data information such as basic information of patient test samples, pathological diagnosis (cancer type), test information (product, hospital, institution, project) and corresponding test results (gene mutation, TMB, MSI); the reporting system is also associated with medical interpretation information, and can obtain various medical interpretation information related to gene mutation; then, the reporting system uses the template creation unit, data management unit and report review unit to automatically interpret the report of the patient test samples, and displays the interpretation content related to the test results of each patient test sample in a standardized form and logic according to the established rules, and generates a PDF / word report with a specific template, which is convenient for the interpreter to review, update, circulate and deliver the automated interpretation report. The present invention can be applied to the automated and precise interpretation of comprehensive tumor gene prompt reports of genetic testing institutions. It has the characteristics of comprehensive and accurate interpretation in accordance with standard guidelines, high degree of automation in the entire process of report collaborative review and delivery, and complete functions. It can complete one-stop multi-link operations in the reporting system, and can support efficient interpretation to issue accurate, comprehensive, professional and authoritative tumor gene testing reports.

[0070] Correspondingly, in other embodiments of the present invention, a method for automatically interpreting and managing tumor gene detection reports is provided.

[0071] Example 1

[0072] like Figure 1As shown, this embodiment provides a system for automatic interpretation and management of tumor gene detection reports, which includes: a template creation unit, a data management unit and a report review unit. Among them, the template creation unit is used to create personalized report templates according to clinical needs; the data management unit is used to obtain upstream data and medical interpretation information of patient test samples based on personalized report templates, generate automated interpretation reports according to preset data processing and interpretation logic, and store report-related data; the report review unit is used to create report review tasks based on the generated automated interpretation reports, and send them to the corresponding review client according to the preset scheduling management rules. The corresponding review client reviews them according to the preset report review management rules, and manages the entire report review process at the same time to obtain the final version of the interpretation report of the patient's test samples.

[0073] Further, if Figure 2 As shown, the template creation unit includes a module management module and a template management module. The module management module is used to establish and store all submodules in the report template, and freely combine different submodules to generate a basic report framework corresponding to different products; the template management module is used to create a template name according to clinical needs and associate it with the basic report framework corresponding to different products, and generate or pre-generate a personalized report template.

[0074] In this embodiment, by setting up two modules, namely a template management module and a module management module, it is possible to create and store multiple personalized report templates that meet different clinical needs at one time, which is simple, fast and efficient to operate.

[0075] Furthermore, in this embodiment, the module management module includes but is not limited to a targeted medication guidance submodule, a typing / prognosis assessment submodule, a variation analysis submodule, an immunotherapy guidance submodule, a genetic risk warning submodule, a chemotherapy drug guidance submodule and a clinical trial submodule.

[0076] Among them, the targeted medication guidance submodule is used to guide targeted medication; the typing / prognosis assessment submodule is used to assist molecular typing and assess prognosis; the variation analysis submodule is used to analyze the biological functions and carcinogenic effects of gene variations in various tumor types; the immunotherapy guidance submodule is used to guide immunotherapy; the genetic risk prompt submodule is used to prompt the germline variation grading and tumor genetic susceptibility risk; the chemotherapy drug guidance submodule is used to guide the use of chemotherapy drugs; and the clinical trial submodule is used to guide the enrollment of tumor patients in clinical trials.

[0077] The module management module is equipped with functions such as create, save, search, delete, and modify, which are used to create, save, search, delete, and modify submodules. For some specific products, if a certain submodule of the product is not available in the module management module, you can directly create and save the submodule through the create function; for product updates and iterations, you can directly search for a submodule, delete or modify the function, re-check the submodule, and form a new report basic framework to achieve the purpose of updating, which can be used to automatically generate reports under different clinical needs.

[0078] Furthermore, the template management module includes a template creation submodule, an associated information submodule and a report generation submodule. Among them, the template creation submodule is used to create template names and number them according to clinical needs; the associated information submodule is used to associate the created template with upstream data, such as product, hospital, institution, project and other testing information; the report generation submodule is used to associate the template name with the report basic framework corresponding to different products created in the module management module to form a personalized report template.

[0079] In this embodiment, several types of personalized report templates can be stored and generated in the template management module in advance, and different products, hospitals, institutions, projects and other information can be combined with corresponding sub-modules to provide personalized report templates for different patient test samples to meet the needs of different scenarios. It can also form a large, medium and small panel product template framework; it can also provide word or PDF version reports to meet the needs of diversified report formats; it can also provide Chinese version, English version, simplified version, standard version reports to meet the needs of diversified report types.

[0080] Furthermore, the data management unit includes a data retrieval module, a data processing module and a data storage module. The data retrieval module is used to obtain the upstream data and medical interpretation information of the patient's test sample according to the personalized report template; the data processing module is used to process the upstream data and medical interpretation information according to the preset data processing and interpretation logic to generate an automated interpretation report; the data storage module is used to store report-related data, including the upstream data and medical interpretation information processed by the data processing module.

[0081] Further, if Figure 3 As shown, the report review unit includes a shift management module, a report management module and a task management module. Among them, the shift management module is used to create a shift schedule according to the preset shift management rules, and automatically assign the report review task to the auditor on duty that day according to the preset allocation principle; the report management module is used for the auditor on duty that day to review the automated interpretation report according to the preset report review management rules and supervise and record the status of the automated interpretation report to obtain the final interpretation report of the patient's test sample; the task management module is used to track the entire process of the report review task in real time.

[0082] Furthermore, the shift management module includes an account and authority setting submodule, a shift schedule creation submodule, and an audit task allocation submodule. Among them, the account and authority setting submodule is used to open corresponding accounts and allocate different permissions for each auditor according to the auditor's work permissions and work content; the shift schedule creation submodule is used to create a shift schedule according to the work permissions and work content of different auditors, and determine the auditors on duty that day; the audit task allocation submodule is used to set audit lines suitable for different scenarios according to different product characteristics, and automatically allocate report audit tasks to auditors on duty that day according to the preset allocation principle.

[0083] Among them, review lines suitable for different scenarios are set according to different product characteristics. For example, products with only target drug modules and products with only genetic variation modules can be specifically assigned to reviewers in different groups. When the report review task is assigned to the reviewer on duty that day, the rule setting principle is random and average distribution.

[0084] Furthermore, the report management module includes an audit submodule and a process management submodule. Among them, the audit submodule is used for each auditor to manually review the automated interpretation report according to the preset report review management rules, and deliver the final interpretation report of the patient test sample after the review is completed; the process management submodule is used to supervise and record the status of the automated interpretation report, and the record content includes patient test sample information (such as report number, report type, terminal customer number, patient name, library number, sample number), product name, institution, template name, cancer type, task type, task rollback type, initial review status, review review status, report generation time, report generation status, sending status, bioinformatics review status and audit path, etc.; in addition, the process management submodule can also confirm the report signatory through the approval flow, and retrieve the corresponding report signatory's signature to the report. At the same time, each report generates a unique version number to distinguish reports generated at different times.

[0085] Furthermore, the task management module includes an online tracking submodule and a report review and statistics submodule. Among them, the online tracking submodule is used to store and view the detailed information of each report review task, and can realize real-time tracking of the entire process of each report review task, including task status, task type, review progress, reviewer, review time, abnormality, etc.; the report review and statistics submodule is used to conduct multi-dimensional review and statistics of genetic testing reports.

[0086] The online tracking submodule traces the complete process of the genetic testing report through relevant notes in the multiple audit circulation process, that is, from the data circulation in the upstream link to the report delivery. Relevant text and documents can be attached and rolled back, so that the report review process is clear, concise and reasonable, and the auditors and important audit information are recorded, solving the accuracy and traceability of the audit data. At the same time, the version of the medical interpretation information associated with the report can be checked.

[0087] The report review and statistics submodule has five functions: report tasks to be processed, report tasks processed, report abnormalities, report termination, and report rollback. Each function can be used to review and count reports through key aspects such as terminal customer number, product, task and review status, reviewer and date. Genetic test reports are automatically stored in the report system, which supports online preview and download within the scope of authority to avoid data leakage and loss and ensure the security of medical data.

[0088] Example 2

[0089] Based on the system for automatically interpreting and managing tumor gene detection reports provided in Example 1, this embodiment provides a method for automatically interpreting and managing tumor gene detection reports, including:

[0090] 1) Create personalized report templates using the template creation unit based on clinical needs;

[0091] 2) Based on the personalized report template, the data management unit is used to obtain upstream data and medical interpretation information, and an automated interpretation report is generated according to the preset data processing and interpretation logic, and the report data is stored;

[0092] 3) The report review unit creates a report review task based on the automated interpretation report, and sends it to the corresponding review client according to the preset scheduling management rules. The corresponding review client reviews the report according to the preset report review management rules, and manages the entire report review process to obtain the final interpretation report of the patient's test sample;

[0093] 4) When the medical interpretation information is updated, the automated interpretation report of the patient's test sample is updated through the data management unit, and the final version of the interpretation report is reviewed again.

[0094] Furthermore, in the above step 1), with the expansion of the genetic testing market and the continuous increase in clinical needs, report templates have more types and personalized requirements. Not only are there differences in report templates for different products, but different clinical projects often have different report template requirements. In the manual issuance mode, report interpreters need to adjust the modules of the report template according to different needs to form the basic framework of the report. This method seriously hinders the work progress of report interpreters and reduces the efficiency and accuracy of interpretation.

[0095] Therefore, this embodiment uses the template creation unit to create a personalized report template, which specifically includes the following steps:

[0096] 1.1) Create corresponding sub-modules in the module management module, and freely combine the sub-modules to generate the basic report framework corresponding to different products.

[0097] Specifically, the steps include:

[0098] First, create a new submodule and fill in the module name, select module category, status and other module information.

[0099] Then, by operating components such as reports, texts, tables, rectangles, and pictures, the required content can be freely combined to obtain sub-modules with specific functions, such as targeted medication guidance, typing / prognosis evaluation, variation analysis, immunotherapy guidance, genetic risk warnings, chemotherapy drug guidance, clinical trials, and other sub-modules.

[0100] In particular, when creating a new submodule, if the new submodule has similar functions to an existing submodule, it can be implemented by copying the existing submodule and then modifying it.

[0101] Finally, different sub-modules are freely combined to obtain the basic reporting framework corresponding to different products.

[0102] 1.2) Create a template name in the template management module and associate test information based on clinical needs.

[0103] Specifically, the steps include:

[0104] First, create a template name and number it in the template management module.

[0105] Secondly, the created template is associated with test information. In this embodiment, the test information includes but is not limited to information such as products, hospitals, institutions, and projects that need to be associated, and these associated information are used to issue a test report corresponding to the template after the sample is tested.

[0106] 1.3) Associate the template name created in the template management module with the report basic framework created in the module management module to generate a personalized report template.

[0107] Furthermore, in the above step 2), the report templates of different products contain different modules, and the data retrieval, data processing and interpretation logic of different modules are also different. Under the traditional manual issuance method, even if the same pathology samples detect the same variant site, there may be differences in the interpretation content of the report issued by different report interpreters. In order to solve the problem of inconsistent and non-standard interpretation of the above reports, for different modules in different report template frameworks, the data retrieval and data processing logic from medical information interpretation to each module in the report is set one by one, so as to realize the automatic and accurate interpretation and standard comprehensive display of report modules such as targeted medication guidance, typing / prognosis evaluation, variation analysis, immunotherapy guidance, clinical trial recommendations, chemotherapy drug guidance, and genetic risk warning in the interpretation report.

[0108] Specifically, the steps include:

[0109] 2.1) Based on the basic information and test information of the patient test samples in the personalized report template, the data retrieval module is used to retrieve pathological diagnosis and test results from upstream data.

[0110] Specifically, the data retrieval module retrieves the upstream data of patient test samples, including basic information of patient test samples (such as name, gender, age, family history, sample type, etc.), test information (such as product, hospital, institution, project), pathological diagnosis (cancer type) and corresponding test results (genetic variation, such as nucleotides, amino acids, exons / introns, transcripts, variation type, variation frequency, etc., TMB, MSI), etc.

[0111] 2.2) Based on data such as pathological diagnosis and test results, the data retrieval module is used to retrieve the corresponding medical interpretation information. At the same time, the data processing module is used to process and interpret the upstream data and medical interpretation information according to the preset data processing and interpretation logic to generate an automated interpretation report.

[0112] Specifically, the steps include:

[0113] 2.2.1) Using pathological diagnosis (cancer type) and test results (gene mutation, TMB, MSI) as input, retrieve the corresponding medical interpretation information according to the preset retrieval logic classification.

[0114] In this embodiment, first, the medical interpretation information to be retrieved is divided into categories such as target drug matching, typing / prognosis, mutation analysis, immunotherapy, chemotherapy drugs, clinical trials, genetic risk analysis, etc. Then, the medical interpretation information is retrieved from each category according to the preset retrieval logic.

[0115] Specifically, the calling logic is:

[0116] ① For target drug matching data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input parameters, and the corresponding targeted drug guidance information is matched in the "Target Drug Matching" category of the medical interpretation information, including targeted drugs, drug evidence, evidence type, sensitivity / resistance, etc., and retrieved to the reporting system.

[0117] ② For classification / prognosis data, the cancer types and genetic mutations corresponding to the patient test samples in the upstream data are used as input parameters, and the classification / prognosis analysis information is matched in the "Classification / Prognosis" category of the medical interpretation information, including prognostic assessment results, evidence, evidence type, etc., and retrieved to the reporting system.

[0118] ③For mutation analysis data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input parameters, and the mutation analysis description information, including the biological function and carcinogenicity of the mutation, is matched in the "Mutation Analysis" category of the medical interpretation information, and retrieved to the reporting system.

[0119] ④ For immunotherapy data, the cancer types and gene mutations, TMB, and MSI corresponding to the patient test samples in the upstream data are used as input parameters, and the analytical information related to the efficacy of immunotherapy is matched in the "immunotherapy" category of the medical interpretation information, including whether the gene mutation has a positive or negative effect on immunotherapy, and retrieved to the reporting system.

[0120] ⑤ For chemotherapy drug guidance data, the cancer type and gene mutation (rs_id and genotype) corresponding to the patient test sample in the upstream data are used as input parameters, and the chemotherapy drug guidance information is matched from the "chemotherapy drug" category of the medical interpretation information, including the toxic and side effects, effectiveness and medication tips of chemotherapy drugs, etc.; retrieved to the reporting system. Among them, rs_id refers to the ID number of the gene locus.

[0121] ⑥ For clinical trial data, the cancer types (including cancer subtypes) and genetic mutations corresponding to the patient test samples in the upstream data are used as input parameters, and clinical trials with completely consistent mutations and cancer types (including cancer subtypes) are matched in the "Clinical Trials" category of the medical interpretation information and retrieved to the reporting system.

[0122] There are tens of thousands of clinical trials related to tumors. How to accurately screen and recommend suitable clinical trials to patients in the report? Relying on manual screening will be very complicated and time-consuming. Currently, most clinical trial recommendations in the interpretation reports on the market are at the gene level or drug level, which is not accurate enough for matching patients. In order to provide patients with corresponding mutations with accurately matched clinical trials, the logic is set: accurate matching mutations, cancer types can be matched to corresponding cancer subtypes, and clinical trials with recruitment status being recruited are displayed in the report. This logic solves the problem of accurately recommending clinical trials suitable for patients; at the same time, the automated system logic can significantly improve the efficiency of report interpretation.

[0123] ⑦ For genetic risk analysis data, the cancer type and test results (gene germline variation) corresponding to the patient test samples in the upstream data are used as input parameters, and the "Genetic Analysis" category of the medical interpretation information is matched with information such as drugs, evidence of medication, type of evidence, sensitivity / resistance, etc., and retrieved to the reporting system.

[0124] Specifically, the gene disease-related descriptions and literature information are obtained from the genetic library-gene disease library through the gene name; the variation classification and classification basis description information of the mutation site are obtained from the genetic library-mutation site database through the gene-nucleotide-amino acid-chromosome-mutation start and end position-reference sequence-variant base recognition retrieval field; the medical interpretation information applicable to the germline pathogenic / potential pathogenic variation is matched in the targeted drug matching, typing prognosis and other modules of the medical interpretation information through the sample cancer type and germline variation pathogenicity label, and the matched drugs, medication evidence, evidence type, sensitivity / resistance and other information are retrieved to the reporting system.

[0125] The automatic identification and acquisition of medical interpretation information can be directly matched to the various modules of the report template, eliminating the need for manual retrieval and entry during the report issuance process, saving a lot of time and energy, and reducing the technical requirements for interpreters, thereby improving report work efficiency and accuracy.

[0126] 2.2.2) The data processing module processes the retrieved medical interpretation information and upstream data according to the preset data processing and interpretation logic, and generates an automated interpretation report according to the personalized report template configuration requirements.

[0127] From medical information interpretation to reporting, related information such as gene mutations and evidence are not all listed in the interpretation report, but need to be flexibly adjusted according to the report display fields of different products and clinical needs. This consumes a lot of time and energy of report interpreters. To solve this problem, this embodiment makes the information display rules and information conversion judgments of each module of the interpretation report into automated logic and stores them in the reporting system.

[0128] The retrieved data such as typing / prognosis, mutation analysis, immunotherapy, chemotherapy drugs, genetic risk analysis, etc. can be automatically filled into the corresponding submodules of the personalized report template after simple logical processing. For target drug matching data and clinical trial data, complex logical processing is required before they are filled into the targeted drug guidance and clinical trial submodules respectively.

[0129] Specifically, when processing target drug matching data, the following steps are included:

[0130] ① According to the preset report display rules and automated judgment logic, determine the evidence level and variation type of target drug information and related evidence in the medical interpretation information.

[0131] In this embodiment, some tags associated with evidence entries in the medical interpretation information are converted to meet the report display requirements to ensure the accuracy of the report interpretation and clinical readability: such as evidence level and variant type.

[0132] Among them, the logic of automatic judgment of evidence level and variant classification is as follows:

[0133] The level of evidence for medical interpretation information is formulated with reference to the "Guidelines for the Interpretation and Reporting of Cancer Sequence Variants" jointly issued by the three authoritative organizations AMP / ASCO / CAP, which divide the level of evidence into four levels: Level A / B / C / D. Evidence from different sources in the medical interpretation information is subdivided according to the type of evidence (status) and accurately matched to the four levels of evidence in the AMP / ASCO / CAP guidelines. The type of evidence approved by the FDA / NMPA or recommended by the NCCN guidelines corresponds to Level A (i.e., Level A evidence); the type of evidence from large-scale clinical studies suggested by expert consensus corresponds to Level B (i.e., Level B evidence); Level A evidence or clinical trial research evidence in other cancers, mainly including published Phase II, Phase III and above clinical research results or ongoing clinical trials, corresponds to Level C (i.e., Level C evidence); case reports and preclinical experimental evidence correspond to Level D (i.e., Level D evidence).

[0134] In this embodiment, the variant types are also divided into three types: I / II / III, and the corresponding relationship between the variant types and the evidence levels is Class I = level A / B; Class II = level C / D; and the others are Class III.

[0135] The level of evidence and variation type should be displayed after each drug recommended in the interpretation report in the form of variation level (I, II, III) - evidence level (A, B, C, D), such as (II-C).

[0136] The "target drug matching"-evidence type value hit by the mutation and cancer type in the medical interpretation information, such as FDA / NMPA, has a corresponding evidence level field value of level A in the logic, and the report output value is "A", and the corresponding post-drug level in the report is displayed as IA; if no relevant entries are hit in the medical interpretation information-target drug matching and medical interpretation information-typing and prognosis modules, it is judged to be a Class III mutation, and the mutation information is displayed in the report Class III mutation module.

[0137] ② Screen target drugs and evidence according to the preset automated logical rules for screening and sorting target drugs and evidence, as well as the determined evidence level and variation type of each piece of evidence.

[0138] There is a lot of data and evidence stored in the medical interpretation information for gene mutations and cancer types. If all of them are retrieved into the interpretation report according to the medical interpretation information, the time for interpretation review will be increased, and the retrieved evidence will not be displayed prominently, resulting in poor report readability. At the same time, different reviewers have different choices for evidence selection, which will also lead to differences in the issuance of similar mutation reports. To solve this problem, in this embodiment, automatic screening and display logic options for report interpretation are set, and automatic target drug and evidence screening and sorting display logic are selected, which can effectively shorten the interpretation time, such as only retrieving and displaying the highest level of evidence for the same mutation and the same drug, thereby improving the accuracy, uniformity, and readability of the interpretation report.

[0139] Display logic of target drugs and evidence ranking: To meet different clinical requirements, it can be flexibly implemented according to needs, such as:

[0140] a. According to the sensitivity or resistance of drugs and evidence, they can be automatically displayed in the report in two sections, with all sensitive drugs and evidence in the front and resistant drugs and evidence in the back;

[0141] b. For the same drug, if the medical interpretation information has high-level evidence, lower-level evidence may not be output;

[0142] c. Evidence of the same drug is displayed together;

[0143] d. Ranking of evidence levels: Level A > level B > level C > level D;

[0144] e. The ranking within the same level is as follows: Level A: FDA / NMPA>NCCN; Level C: clinical trial period from high to low (Phase 4>Phase 3>Phase 2 / 3>Phase 2>Phase 1 / 2)>review studies, etc.;

[0145] f. When encountering contradictory evidence for the same drug, the target drug matching data can be sorted according to the level of the evidence itself, such as sensitive evidence in the sensitive section and drug resistance evidence in the drug resistance section.

[0146] ③ Process the screened target drugs and evidence according to the preset automated logic of conflicting evidence judgment, level determination and display.

[0147] When interpreting reports manually, a common and time-consuming scenario is when a variant is detected in the report, and the related drug benefit prompts have both sensitive evidence and drug resistance evidence. For complex and contradictory evidence, in addition to considering the content and level of each clinical evidence, the report interpretation team needs to discuss and determine the final drug recommendation, evidence presentation, and level determination details. Discussions and opinions reached at different times or by different report interpreters may be subjective, leading to differences in the interpretation of similar reports. Therefore, in order to solve the problem of difficult and time-consuming judgment of contradictory evidence, automated logic for the judgment, level determination, and presentation of contradictory evidence in different scenarios has been developed.

[0148] Set options for automatic judgment and display logic of complex contradictory evidence: When retrieving medical interpretation information to the reporting system to produce reports, if there are multiple contradictory evidences of the same variant and the same drug, the automatic judgment logic can be triggered according to the preset rules, including judging the priority sorting and display of evidence; for example, when the no-benefit evidence only appears at the same time as the drug-resistant evidence, and there is no other sensitive type of evidence, it can be decided whether to retrieve and display the no-benefit evidence according to needs.

[0149] When processing clinical trial data, this includes:

[0150] According to clinical needs, clinical trials can be further automatically screened and sorted. For example, only clinical trials with a recruitment status of recruiting can be displayed, and they can be sorted according to clinical stages: Phase 4 > Phase 3 > Phase 2 / 3 > Phase 2 > Phase 1 / 2, etc.

[0151] Furthermore, the above step 3) includes the following steps:

[0152] 3.1) Use the shift management module to create a shift schedule and automatically assign the automated interpretation report to the reviewer on duty that day according to the preset allocation principle.

[0153] After the data management unit retrieves, processes, and stores the upstream data and medical interpretation information, it generates automated interpretation reports for different needs. After the shift management module receives the automated interpretation reports for different needs, it distributes the automated interpretation reports to the auditors on duty that day according to the shift schedule and product, cancer type information, and mutation information, in accordance with the principle of random and average distribution. The automated distribution allows each auditor to randomly receive a different automated interpretation report. In particular, if an auditor encounters a report with a large workload / difficulty, he or she can manually assign the report to other auditors, and reasonably configure it to improve work efficiency.

[0154] 3.2) The corresponding reviewers review the automated interpretation report according to the preset review process to obtain the final interpretation report of the patient's test sample, and track the entire report review process in real time through the task management module.

[0155] In this embodiment, the audit process set includes:

[0156] ① Two-person review mechanism

[0157] In order to ensure the professionalism and accuracy of the report content, this report system has set up a two-person review mechanism, which divides the reviewers into initial reviewers and reviewers. After the initial reviewer completes the report review, the report cannot be delivered directly, and it needs to be delivered to the reviewer for further review. Only after the reviewer's review can the "Send" option be clicked in the report system to deliver the report.

[0158] ②All report review tasks can be transferred in both directions

[0159] During the report review task flow, it can be transferred to the previous stage or the next stage in real time as needed. For example, the reviewer can judge whether to return the report review task to the reviewer for further modification or to the next stage based on the review result of the initial reviewer; if data problems are found during the report review, they can be returned to the upstream stage for data modification; if problems are found in the automatic interpretation of the report in the downstream stage, they can be returned to the report review stage for modification.

[0160] ③Set different workflows according to different needs

[0161] Responsible persons in different fields are classified into different workflows, and one or more responsible persons can be set in the workflow according to actual needs. The review order of the responsible persons is set according to the rules. When all the workflows configured for the task are completed, the last responsible person confirms and completes the work. If different products or types of cancer need to flow through different workflows, the system will configure the required workflows according to the product information. After the reviewers of each workflow are completed, the last reviewer will deliver the report. The multiple review mechanism can ensure the accuracy of the report. In addition, the report approval flow can confirm the report signer and retrieve the corresponding report signer's signature to the report. At the same time, each report generates a version number with a unique value to distinguish reports generated at different times.

[0162] ④Set notes

[0163] In the process of multiple audits, the relevant notes are optional and fillable. Common notes are set as options. One-click selection can efficiently complete the audit information and transfer it to the subsequent stage. Relevant text, pictures and files can be attached to achieve a clear, concise and reasonable report audit process. The auditor and important audit information are recorded and left with traces, solving the accuracy and traceability of audit data. The online browsing system generates files to ensure the accuracy of the files and prevent problems caused by unexpected events such as format errors or file transmission errors during the transmission process.

[0164] Report delivery involves not only communication and task allocation among personnel in the same department, but also cross-departmental collaboration. Once communication is not timely and task allocation is unreasonable, the circulation efficiency will be reduced and timely delivery will not be possible. The reporting system conducts online review of genetic test reports and sends and delivers the approved reports, which can be efficiently circulated, shorten the time and ensure timeliness. While reducing manpower and time costs and optimizing resource allocation, it also improves the accuracy and timeliness of the test reports.

[0165] After the auditor has reviewed the interpretation report for multiple times and passed it, click "Send" to complete the report delivery. After the interpretation report is delivered, if the data is updated due to different reasons, it can be transferred to the report system again to complete the second delivery.

[0166] Preferably, in the above step 4), the auditor may encounter the situation that the latest clinical evidence is not interpreted in time during the process of interpreting and reviewing the report. The reason may be that the evidence of the medical interpretation information has not been sorted and updated in time; it may also be that some clinical trial evidence has been stopped due to poor efficacy. Therefore, it is necessary to synchronously update the medical interpretation information and the interpretation report. In this embodiment, the method for synchronously updating the medical interpretation information and the interpretation report is:

[0167] 4.1) Set up the report data function in the report management module, which can be linked to the data management unit;

[0168] 4.2) A data update function is set in the data management unit, which can retrieve, process and store the medical interpretation information according to the preset retrieval logic, and update the automated interpretation report.

[0169] Based on the report data function and data update function, it is not necessary to process and update each similar report one by one. It is only necessary to complete the addition or deletion of the evidence in the medical interpretation information once, and then trigger the report data function in the report management module of the reporting system, which will be linked to the data management unit, and then trigger the data update function to obtain data, execute the preset data processing logic, and complete the update and adjustment of the automated interpretation report. Therefore, when manually reviewing the PDF report automatically generated by the reporting system, if the information that needs to be adjusted is found, such as mutation analysis, recommended drug level and evidence analysis addition, deletion or modification, etc., it can be directly operated in the medical interpretation information to complete the update of the medical interpretation information; then trigger the report data function for the report in the reporting system, link to the data management trigger acquisition data function, execute the data processing script, regenerate the report, and complete the update of the report interpretation. On the one hand, it saves the time of modifying each report one by one and improves the efficiency of report interpretation and review; on the other hand, directly modifying the medical interpretation information to complete the update of the report is more accurate and reduces the introduction of human errors.

[0170] Example 3

[0171] This embodiment uses a specific case to further introduce the automated interpretation and management system for tumor gene detection reports proposed by the present invention.

[0172] For example, when hospital H1 sends patient sample S1 for testing of product P1, S1 starts from the sample collection and recording, to the wet experiment, and then to the bioinformatics analysis of the dry experiment. These processes are transferred and stored upstream in the reporting system. In the reporting stage of the dry experiment, the reporting system begins to be used. The data retrieval module of the reporting system is connected with the upstream data and medical interpretation information. After matching the personalized report template of the patient's test sample through the module management module and the template management module, the logic is set to enable the reporting system to retrieve the upstream data, including the basic information, cancer type, test information, test results, etc. of the patient sample S1; then, according to the sample pathological diagnosis (cancer type) and test results (gene variation, including somatic variation and germline variation, TMB, MSI, etc.) information, the corresponding report sub-module medical interpretation information is classified and matched from the medical interpretation information, where the medical interpretation information includes target drug matching, typing / prognosis, mutation analysis, immunotherapy, genetic analysis, chemotherapy drugs, clinical trials, references, etc.; the corresponding report generally includes sub-modules such as targeted drug guidance, typing / prognosis evaluation, variation analysis, immunotherapy guidance, genetic risk warning, chemotherapy drug guidance, clinical trials, etc.; according to the hospital H1 and product S1, the template management module matches the corresponding template, and then the upstream data and medical interpretation information are integrated and processed through the data processing module, and stored using the data storage module to automatically generate a report for sample S1 ( Figure 1), and then store the generated report in the task management module, and reasonably assign tasks to corresponding report reviewers according to the shift management module. The report reviewers review the report in the report management module, minimizing unnecessary manual participation in the work, which is fast and efficient.

[0173] Specifically, the functions of each module are introduced as follows:

[0174] 1) Template creation unit - create personalized templates

[0175] Create a template name. For example, if the H1 hospital sends tissue sample S1 for testing of product P1 and product P2, before the gene test report of the patient's test sample is issued, it is necessary to create a template name in the template management module of the report system and number it, such as H1P1_001 and H1P2_001 respectively; then select the template-related information, such as the associated product, hospital, institution, project and other information. This associated information is used to issue the report of the corresponding template after the sample is tested.

[0176] Submodule combination. If the submodules in the H1P1_001 template are all submodules already in the reporting system module, you can add the required submodules to the H1P1_001 template and save them by direct search or classification search, thus completing the template configuration of the H1 hospital P1 product. For example, if the product requires targeted medication guidance, typing / prognosis evaluation, variation analysis, immunotherapy guidance, genetic risk warning, chemotherapy drug guidance, clinical trials and other submodules, you can check the submodules that the product should include in the module management module, generate the basic report framework of the product, and then complete the configuration of the template name and related information, and associate the template name with the product in the template management module.

[0177] If some sub-modules in the report template required by H1P2_001 are not included in the module management module, you can create the required sub-modules in the new function of the module management module, save them, and then continue to check to generate the basic report framework of H1P2_001. If a product has sub-modules that need to be updated and iterated, you can directly search for the sub-module name in the module management module to modify or delete it. In this embodiment, the generated personalized report template will be stored in the template management module, and the created personalized report template can be modified. If there is a modification, the system will automatically generate a version number for the personalized report template, such as H1P1_001 The modified version number is H1P1_001_v202307201234, and the data management unit can view the template version number used when generating the automated report. The H1P1_001 and H1P2_001 report templates are used for the automated generation of reports for the S1 sample P1 product and P2 product of the data management unit.

[0178] (2) Data management unit - data retrieval, processing and storage

[0179] The upstream test data obtained by the data retrieval module of the data management unit include basic information of patient test samples (such as name, gender, age, family history, sample type, etc.), test information (such as product, hospital, institution, project), pathological diagnosis (cancer type) and corresponding test results (genetic variation, such as nucleotides, amino acids, exons / introns, transcripts, variation type, variation frequency, etc., TMB, MSI), etc.

[0180] The data retrieval module is used to retrieve different medical interpretation information according to the retrieved upstream data and the created report template, and the information is interpreted according to the preset processing logic of the data processing module, and the data storage module is used to store the data.

[0181] For target drug guidance data, for example, if the upstream data shows that the cancer type of the patient's test sample is non-small cell lung cancer, and the gene mutation is EGFR p.L858R, this is used as the input information. According to the set interpretation rule logic, the medical interpretation information is automatically matched with the clinical information one by one, and the target drug matching results are output to the targeted drug guidance submodule in the report: "Target drug matching" information matches the drugs in the target drug guidance: "gefitinib, erlotinib, dacomitinib, osimertinib, erlotinib combined with ramucirumab, osimertinib, ametinib, vometinib, afatinib, icotinib, etc.; matching drug mutations and evidence levels: IA; matching drug evidence: "FDA approved gefitinib, erlotinib, dacomitinib, osimertinib, erlotinib combined with ramucirumab for the treatment of non-small cell lung cancer patients with EGFR gene exon 19 deletion or EGFR-p.L858R; NMPA approved gefitinib, erlotinib, dacomitinib, osimertinib, ametinib, and vometinib for the treatment of patients with non-small cell lung cancer with EGFR gene exon 19 deletion or EGFR-p.L858R"; "FDA approved afatinib for the treatment of patients with non-small cell lung cancer with EGFR non-resistant mutations; NMPA approved afatinib for the treatment of patients with non-small cell lung cancer with EGFR gene sensitive mutations"; "NMPA approved icotinib for the treatment of patients with non-small cell lung cancer with EGFR gene sensitive mutations"; matching evidence type: "FDA / NMPA"; sensitive / resistant: "sensitive" and other information.

[0182] For classification / prognosis data, for example, upstream data shows that the cancer type of the patient's test sample is a central nervous system tumor, and the gene mutation detected is a PRKCA p.D463H missense mutation. This is used as the input information, and according to the set interpretation rule logic, the medical interpretation information is automatically matched one by one with the clinical medical interpretation information, and then to the classification / prognosis evaluation submodule in the report to output the classification / prognosis result: The "classification / prognosis" information matches the prognosis evaluation result, that is, the auxiliary molecular diagnosis result is "glioma"; the evidence matching the prognosis evaluation result: "WHO Central Nervous System Tumors Classification Guidelines (WHO CNS5) indicate that the PRKCA gene p.D463H missense mutation is the molecular characteristic of "chordal glioma"; matching evidence type: "WHO".

[0183] For example, in a clinical test sample, the cancer type is non-small cell lung cancer, and the mutation detected is a KRAS p.G12D missense mutation; this is used as the input information, and the medical interpretation information is automatically matched with the clinical information one by one according to the set interpretation rule logic, and the classification / prognosis evaluation submodule in the report outputs the classification / prognosis result: Prognosis evaluation result and evidence: "NCCN's "Clinical Practice Guidelines for Non-Small Cell Lung Cancer" suggests that patients with detected KRAS mutations have a worse survival prognosis than those without detected KRAS mutations", and the evidence type: "NCCN".

[0184] For mutation analysis data, such as a patient test sample with non-small cell lung cancer, the mutation detected was EGFR p.L858R missense mutation. This is used as the input information. According to the set interpretation rule logic, the "mutation analysis" information in the medical interpretation information is matched to the mutation analysis submodule in the report, and the mutation analysis information is output: "EGFR p.L858R mutation is a common missense mutation in exon 21, located in the tyrosine kinase domain of EGFR. Non-small cell lung cancer (NSCLC) patients carrying EGFR p.L858R mutation are sensitive to EGFR tyrosine kinase inhibitors (EGFR-TKIs)".

[0185] For immunotherapy data, such as a clinical test sample, the cancer type is cervical cancer, and the immune marker detected is TMB-H; this is used as the input information, and the medical interpretation information is automatically matched to the clinical information one by one according to the set interpretation rule logic; the "immunotherapy" information of the medical interpretation information is matched to the report immunotherapy guidance submodule: output immunotherapy effect analysis: "Currently, the FDA has approved pembrolizumab for the treatment of patients with unresectable or metastatic solid tumors with high tumor mutation burden (TMB-H, TMB ≥ 10muts / Mb), disease progression after previous treatment, and no other preferred treatment options (Note: The safety and efficacy of pembrolizumab in children with TMB-H central nervous system tumors has not yet been determined). The NCCN "Clinical Practice Guidelines for Cervical Cancer" recommends the use of pembrolizumab as a second-line or subsequent treatment option for patients with TMB-H tumors or PD-L1-positive or MSI-H / dMMR recurrent or metastatic cervical squamous cell carcinoma, adenocarcinoma, or adenosquamous carcinoma."

[0186] For chemotherapy drug guidance data, for example, a clinical test sample, the cancer type is colorectal cancer; the detected gene variation information (chemotherapy site type) is the DPYD gene rs55886062 AC heterozygous mutation; this is used as the input information, and the medical interpretation information is automatically matched one by one with the clinical information according to the set interpretation rule logic; the "chemotherapy drug" information of the medical interpretation information is matched to the reported chemotherapy drug guidance submodule, and the chemotherapy guidance effect analysis is input: "Patients with colorectal cancer carrying the AC genotype at the DPYD gene rs55886062 site may have a higher risk of drug toxicity compared to the AA genotype", and drugs: chemotherapy regimens based on fluorouracil drugs.

[0187] For clinical trial data, for example, if upstream data shows that the mutation is EGFR p.L858R and the cancer type is non-small cell lung cancer, go to Medical Interpretation Information-"Clinical Trials" to find all clinical trials that are recruiting for EGFR p.L858R and non-small cell lung cancer. You can sort them comprehensively based on tags such as mutation, cancer type, and period, and put clinical trials with higher relevance in the front. Match the clinical trial submodule in the report and output the clinical trial results: Drug: AZD3759; Recruitment status: Recruiting; Trial stage: Phase 3; Recruitment location: China.

[0188] For genetic risk analysis data, for example, the gene description retrieved through BRCA1: BRCA1 gene is an important tumor suppressor gene, and its encoded protein participates in DNA damage repair, and also plays an important role in cell cycle regulation, gene transcription regulation and induction of apoptosis. Certain inherited mutations of BRCA1 gene are associated with the risk of developing hereditary breast and ovarian cancer syndrome (HBOC) with autosomal dominant inheritance. Patients with this syndrome have an increased risk of breast cancer and ovarian cancer (including fallopian tube cancer and primary peritoneal cancer). The cumulative risk of breast cancer and ovarian cancer in individuals carrying BRCA1 gene mutations at the age of 70 increases to 44%-78% and 18%-54%, respectively; patients with this syndrome also have an increased risk of other tumors compared with the general population, such as pancreatic cancer, prostate cancer, biliary tract cancer, gastric cancer, lymphoma and lung cancer. In addition, the BRCA1 gene is also associated with the risk of Fanconi Anemia (FA), a rare genetic disease with diverse clinical manifestations, characterized by congenital physical malformations, bone marrow hematopoietic disorders, and susceptibility to tumors, including hematological malignancies and non-hematological malignant solid tumors, such as myelodysplastic syndrome, acute lymphoblastic leukemia, head and neck squamous cell carcinoma, etc. BRCA1 gene-related Fanconi anemia is inherited in an autosomal recessive pattern. Patients with this syndrome often have homozygous or compound heterozygous pathogenic mutations in related genes (i.e., pathogenic mutations appear in both copies of the gene), and heterozygotes usually do not show signs and symptoms of related diseases. The variation grade and grade basis of the mutation site are obtained from the medical interpretation information-"Genetic Analysis Data" through the fields of gene-nucleotide-amino acid-chromosome-mutation start and end position-reference sequence-variant base recognition. For example, through BRCA1-c.2975_2979del-17-41244569-41244573-TTTAG-the grade can be uniquely matched from the mutation site: pathogenicity, grade basis: The BRCA1 p.Thr992MetfsTer2 mutation you carry is not included in the ClinVar database; this mutation is not included in the population databases 1000_CN, 1000_MAF, ESP6500, and gnomAD, or the frequency is extremely low; the loss-of-function variation of the BRCA1 gene is the pathogenic mechanism of the disease; this mutation will terminate protein translation prematurely, and the function of the resulting truncated protein may be affected; in summary, this mutation is determined to be a pathogenic mutation. This suggests that other family members (such as parents, children, siblings, etc.) are likely to also carry the mutation. It is recommended that the family members of the tested person undergo relevant genetic testing and implement early preventive measures.

[0189] In addition, the variation classification and classification basis of the mutation sites identified and retrieved through the data retrieval module will also be displayed in the GeneTron_Classification and GeneTron_Result description columns of the reporting system-offline data-variation type germline interface, and the mutation list information will be displayed in the order of variation classification, which are: empty (that is, the genetic interpretation information does not include the site, and the interpretation information is not retrieved), pathogenic, possibly pathogenic, uncertain, possibly benign, benign. Medical interpretation information-sites not included in the genetic analysis will be displayed at the top of the mutation list to facilitate report reviewers to identify unincluded genetic variations, and import the genetic interpretation information after interpretation for report issuance. Report reviewers also review the reporting system offline data-variation type germline interface information and report display information. If the report information is correct, there is no need to modify and update the medical interpretation information, and it can be directly transferred and delivered for automatic report generation.

[0190] 3) Report review unit - report review process

[0191] 3.1) Report review and review process

[0192] There are different testing requirements for different testing products, such as testing genetic susceptibility-related gene mutations, testing tumor drug-related sensitivity or MRD testing, etc. Some of these different requirements only require genetic line or drug line review, and some require genetic line and drug line review at the same time. Based on this, two parallel review lines are set up in the reporting system. When a report A1 containing drug sensitivity detection and genetic susceptibility detection flows from the upstream to the reporting system, combined with the product information of A1 entered in the upstream data, the reporting system automatically configures two review lines, the drug line and the genetic line. After the initial review and re-review of the genetic line are completed, it will be transferred to the initial review of the drug line, and then to the re-review of the drug line. Usually, the report management module has two steps: initial review and re-review. When the review finds an abnormality, if it is an upstream abnormality, the task can be rolled back to the upstream; if it is an initial review abnormality, the task can be rolled back to the report management module of the initial reviewer.

[0193] When abnormal mutation sites are found in the report during the review process and need to be investigated, the reviewer has the authority to roll back the investigation and can write the reason in the remarks and return it to the relevant upstream person in charge. After the upstream person in charge confirms the problem, it will be transferred back to the report issuance process. This process enables the rapid implementation of a dedicated person responsible system.

[0194] 3.2) Report delivery

[0195] The order recorder enters the basic customer information upstream as follows: terminal customer number: A; name: B; product: C; hospital: D; report receiving email: E; cancer type: non-small cell lung cancer. After the data analyst completes the data, the report system automatically generates two WORD reports, a simplified version and a standard version, according to customer needs. After the auditor double-checks and passes, the report system clicks "Send" to complete the report delivery. After the MDT multidisciplinary consultation, it is required to change the sample pathology from non-small cell lung cancer to small cell lung cancer. First, the upstream data is submitted for modification. After the modification is completed, the report system automatically regenerates the small cell lung cancer report, and the report system reviews and delivers it again.

[0196] 3.3) Report Archiving and Data Security

[0197] The reporting system tracks the genetic test report online, and can trace the complete report process. The relevant notes during the multiple review and transfer process can be viewed from the data analyst to the report system to generate the report and the final report sending process. For example, the information of a report is as follows: terminal customer number: A; name: B; product: C; hospital: D; report receiving email: E; cancer type: non-small cell lung cancer. The report system notes first record the data analysis personnel F's data transfer online, and the system automatically triggers the generation of the report, then records the double review of the report interpretation personnel G and the medication personnel H, and finally records the medication personnel G clicks to send to complete the report sending. When reviewing the report, if the medical interpretation information is modified, the data needs to be re-acquired, and the report system will also record the version of the medical interpretation information associated with the report. After the report is delivered, the upstream data modifies the cancer type, and the reporting system records the cause of the disease from non-small cell lung cancer to small cell lung cancer. The report task type that is re-transferred to the reporting system is changed from normal to rollback, and it needs to be reviewed again. The flow process is the same as the initial review, and the relevant notes are also recorded.

[0198] 3.4) Full process tracking of report review tasks

[0199] At present, the workflow and data of automated report interpretation of tumor gene testing are difficult to trace, and the report review and statistics are of single dimension, difficult to store, and have a low safety factor. The task management of the reporting system can track the gene testing report online, which not only realizes multi-dimensional report review and statistics, but also can automatically store any issued gene testing report in the reporting system and associate it with the data storage module to avoid data leakage and loss, and ensure the data security of the company and its customers.

[0200] The task management module of the reporting system tracks the genetic testing reports online. The relevant notes during the multiple review processes can trace the entire report process. It can be previewed and downloaded online to avoid data leakage and loss, and ensure the data security of the company and its customers.

[0201] The task management module is used to view the detailed information of each task and to track the entire process of the task, including task status, task type, review progress, reviewer, review time, and whether there are any abnormalities.

[0202] This embodiment is further introduced with a specific case to illustrate the specific scenarios of system functions and the relationship between modules.

[0203] In the task management module of the reporting system, by inputting the terminal customer number A, you can confirm that the report with this number has been processed and is a rollback task. At the same time, you can view the customer name, sample, library, product, cancer type and review status online, and view the notes. Not only can you trace the complete process of sending the report for the first and second review, but you can also obtain the rollback reason and the medical interpretation information version associated with the report. At the same time, you can also preview and download the report sent for review online. In addition, you can also input multiple keywords such as product C, cancer type, date, and auditor at the same time to count the specific types of reports in the pending, processed, abnormal, terminated and rollback modules, and realize report review and statistics in multiple dimensions. The genetic test report of any module can not only be previewed and downloaded online, but also automatically stored in the reporting system to avoid data leakage and loss, and ensure the data security of the company and customers.

[0204] 4) Interpretation reports and medical interpretation information are updated simultaneously

[0205] For example, during the review of the report, a non-small cell lung cancer sample was found with a p.L858R mutation. The report needs to supplement the BLU-945 combined with osimertinib drug reported in 2023 AACR and its related preclinical evidence. At this time, there is no need to operate the PDF report. You can directly add the target drug BLU-945 combined with osimertinib in the target drug matching module of the medical interpretation information, and the evidence "The 2023 AACR report, the EGFR-p.L858R BaF3 xenograft model showed that compared with osimertinib alone, BLU-945 combined with osimertinib prolonged the duration of response. The SYMPHONY (NCT04862780) clinical trial is studying the combination of BLU-945 + osimertinib to treat EGFR-p.L858R non-small cell lung cancer" and other information. After the medical interpretation information is updated and published, search for the report number in the report system, click Re-acquire Data, and generate the latest report. This completes adding new drugs to the report. Subsequent similar reports can be completed by modifying the medical interpretation information once, without the need to perform operations on each report.

[0206] In addition, when the general interpretation information obtained in the formal medical interpretation information-genetic analysis is not suitable for the personalized issuance of a special report, such as under normal circumstances, the genetic part of the report of large panel and other drug products only reports pathogenic / potentially pathogenic level mutations, and the other unknown / potentially benign / benign levels are not reported; in special circumstances, reports that are different from the above rules may also be encountered, such as the customer's request to report unknown level mutations in a pure negative report, a hospital project has special requirements not to report pathogenic / potentially pathogenic mutations of specific genes, adjustments to the reporting of pathogenic / potentially pathogenic mutations outside the product scope, adjustments to the level of genetic results reported for a patient at different times, and the need to display the level adjustment and the reasons in the report, etc. Personalized report issuance can be achieved through temporary mutation site data and temporary gene disease number functions. The terminal customer number, sample number, product ID, gene, mutation and interpretation information are filled in the temporary mutation site data and / or temporary gene disease data. When the reporting system obtains the medical interpretation information, it will give priority to retrieving the temporary interpretation information for the issuance of a personalized report for the specific sample, avoiding interference with other reports caused by the modification and reversion of the formal medical interpretation information. For example, a product sample is reported to increase personalized grading basis analysis information for the heterozygous variant site of the SBDS c.258+2T>C autosomal recessive genetic disease detected at the request of the customer. After a strict application process approval, the report interpreter enters the patient's terminal customer number, sample number, test item ID, mutation site, grade and personalized grading basis analysis in the temporary mutation site of the medical interpretation information. When generating the report, the sample analysis will give priority to retrieving the temporary medical information to generate a personalized report that meets customer needs.

[0207] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A system for automatically interpreting and managing tumor gene detection reports, characterized in that: include: Template creation unit, data management unit and report review unit; The template creation unit is used to create a personalized report template according to clinical needs; The data management unit is used to obtain upstream data and medical interpretation information of patient test samples based on a personalized report template, generate an automated interpretation report according to preset data processing and interpretation logic, and store report-related data; The report review unit is used to create a report review task according to the generated automated interpretation report, and send it to the corresponding review client according to the preset scheduling management rules. The corresponding review client reviews it according to the preset report review management rules, and manages the entire process of the report review task to obtain the final interpretation report of the patient's test sample; The data management unit includes a data retrieval module, a data processing module and a data storage module; the data retrieval module is used to obtain upstream data and medical interpretation information of patient test samples according to a personalized report template; the data processing module is used to process and interpret the upstream data and medical interpretation information according to a preset data processing and interpretation logic to generate an automated interpretation report; the data storage module is used to store report-related data, including the upstream data and medical interpretation information processed by the data processing module; The processing and interpretation of upstream data and medical interpretation information according to the preset data processing and interpretation logic to generate an automated interpretation report includes: Using pathological diagnosis and test results as input, retrieve corresponding medical interpretation information according to preset retrieval logic classification; Among them, the preset calling logic is: ① For target drug matching data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input, and the corresponding targeted drug guidance information is matched in the "target drug matching" category of the medical interpretation information and retrieved into the system, including targeted drugs, drug evidence, evidence type, and sensitivity / resistance; ② For classification / prognosis data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input parameters, and the classification / prognosis analysis information is matched in the "classification / prognosis" category of the medical interpretation information and retrieved into the system, including the prognosis assessment results, evidence, and evidence type; ③ For mutation analysis data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input parameters, and the mutation analysis description information is matched in the "Mutation Analysis" category of the medical interpretation information and retrieved into the system, including the biological function and carcinogenicity of the mutation; ④ For immunotherapy data, the cancer types and gene mutations, TMB, and MSI corresponding to the patient test samples in the upstream data are used as input parameters, and the immunotherapy efficacy-related analytical information is matched in the "immunotherapy" category of the medical interpretation information and retrieved into the system, including whether the gene mutation has a positive or negative effect on immunotherapy; ⑤ For chemotherapy drug guidance data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input, and the chemotherapy drug guidance information is matched from the "chemotherapy drugs" category of the medical interpretation information and retrieved into the system, including the toxic and side effects, effectiveness and medication tips of chemotherapy drugs; ⑥ For clinical trial data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input parameters, and the clinical trial data with completely consistent mutations, cancer types and cancer subtypes are matched in the "clinical trial" category of the medical interpretation information and retrieved into the system; ⑦ For genetic risk analysis data, the cancer types and gene germline variations corresponding to the patient test samples in the upstream data are used as input parameters, and the drugs, drug use evidence, genetic risk analysis, evidence type, and sensitive / resistant information are matched in the "Genetic Analysis" category of the medical interpretation information and retrieved into the system; Process the retrieved medical interpretation information and upstream data according to the preset data processing and interpretation logic, and generate an automated interpretation report according to the personalized report template configuration requirements; Among them, the generation of automated interpretation reports includes: ① The retrieved data on typing / prognosis, mutation analysis, immunotherapy, chemotherapy drugs, and genetic risk analysis are automatically filled into the corresponding submodules of the personalized report template after processing; ② For target drug matching data and clinical trial data, they are processed according to the preset data processing and interpretation logic and then filled into the targeted drug guidance submodule and clinical trial submodule respectively; Among them, when processing target drug matching data, it includes: Determine the evidence level and variation type of target drug information and related evidence in the medical interpretation information based on preset report display rules and automated judgment logic; Screen target drugs and evidence according to preset automated logic rules for screening and sorting target drugs and evidence, as well as the determined evidence level and variation type of each piece of evidence; The preset automated logic rules for drug targets and evidence screening and sorting include: a. According to the sensitivity or resistance of drugs and evidence, they are automatically displayed in the report in two sections, with all sensitive drugs and evidence in the front and resistant drugs and evidence in the back; b. For the same drug, when the medical interpretation information has high-level evidence, the lower-level evidence will not be output; c. Evidence for the same drug is presented together; d. Sort different evidence by level; e. Evidence of the same level is sorted by variant type; clinical trial phases are arranged from high to low; f. When encountering contradictory evidence for the same drug, the evidence for target drug analysis shall be ranked according to the level of the evidence itself; According to the preset automatic logic of conflicting evidence judgment, level determination and display, the selected target drugs and evidence are further screened and sorted; The report review unit includes a shift management module, a report management module and a task management module; the shift management module is used to create a shift schedule according to the preset shift management rules, and automatically assign the report review task to the auditor on duty that day according to the preset allocation principle; the report management module is used for the auditor on duty that day to review the automated interpretation report according to the preset report review management rules and supervise and record the status of the automated interpretation report to obtain the final interpretation report of the patient's test sample; the task management module is used to track the entire process of the report review task in real time; The shift management module includes an account and authority setting submodule, a shift schedule creation submodule and an audit task allocation submodule; the account and authority setting submodule is used to open corresponding accounts and allocate different permissions for each auditor according to the auditor's work authority and work content; the shift schedule creation submodule is used to create a shift schedule according to the work authority and work content of different auditors, and determine the auditors on duty on that day; the audit task allocation submodule is used to set audit lines suitable for different scenarios according to different product characteristics, and automatically allocate report audit tasks to auditors on duty on that day according to preset allocation principles; The report management module includes an audit submodule and a process management submodule; the audit submodule is used for each auditor to manually review the automated interpretation report according to the preset report audit management rules, and deliver the final version of the interpretation report of the patient test sample after the audit is completed; the process management submodule is used to supervise and record the status of the automated interpretation report, and the record content includes patient test sample information, product name, institution, template name, cancer type, task type, task rollback type, initial review status, review review status, report generation time, report generation status, sending status, bioinformatics review status and audit path; The task management module includes an online tracking submodule and a report review and statistics submodule; the online tracking submodule is used to store and view detailed information of each report review task, and to track the entire process of each report review task in real time, including task status, task type, review progress, reviewer, review time, and whether there is an abnormality; the report review and statistics submodule is used to perform multi-dimensional review and statistics on genetic testing reports in the system.

2. A system for automatically interpreting and managing tumor gene detection reports as claimed in claim 1, characterized in that: The template creation unit includes a module management module and a template management module; The module management module is used to establish and store all sub-modules in the report template, and freely combine different sub-modules to generate a basic report framework corresponding to different products; The template management module is used to create template names according to clinical needs and associate them with the report basic frameworks corresponding to different products, and generate or pre-generate personalized report templates.

3. A system for automatically interpreting and managing tumor gene detection reports as claimed in claim 2, characterized in that: The module management module at least includes a targeted medication guidance submodule, a typing / prognosis evaluation submodule, a variation analysis submodule, an immunotherapy guidance submodule, a genetic risk prompting submodule, a chemotherapy drug guidance submodule and a clinical trial submodule; the targeted medication guidance submodule is used to guide targeted medication; the typing / prognosis evaluation submodule is used to assist molecular typing and evaluate prognosis; the variation analysis submodule is used to analyze the biological functions and carcinogenic effects of gene mutations in various tumor types; the immunotherapy guidance submodule is used to guide immunotherapy; the genetic risk prompting submodule is used to indicate germline variation grading and tumor genetic susceptibility risk; the chemotherapy drug guidance submodule is used to guide the use of chemotherapy drugs; the clinical trial submodule is used to guide the enrollment of tumor patients in clinical trials; The template management module includes a template creation submodule, an association information submodule and a report generation submodule; the template creation submodule is used to create a template name and number it according to clinical needs; the association information submodule is used to associate upstream data with the created template; the report generation submodule is used to associate the template name and the report basic framework corresponding to different products created in the module management module to form a personalized report template.

4. A method for automatically interpreting and managing tumor gene detection reports using the system according to any one of claims 1 to 3, characterized in that: The following steps are involved: 1) Create personalized report templates using the template creation unit based on clinical needs; 2) Based on the personalized report template, the data management unit is used to obtain the upstream data and medical interpretation information of the patient's test samples, and an automated interpretation report is generated according to the preset data processing and interpretation logic, and the report data is stored; 3) The report review unit creates a report review task based on the automated interpretation report, and sends it to the corresponding review client according to the preset scheduling management rules. The corresponding review client reviews the report according to the preset report review management rules, and manages the entire process of the report review task to obtain the final interpretation report of the patient's test sample.

5. The method for automatic interpretation and management of tumor gene detection reports according to claim 4, characterized in that: In the step 1), according to clinical needs, a personalized report template is created using a template creation unit, including the following steps: 1.1) Create corresponding sub-modules in the module management module, and freely combine the sub-modules to generate the basic report framework corresponding to different products; 1.2) Create a template name in the template management module and associate test information according to clinical needs; 1.3) Associate the template name created in the template management module with the report basic framework created in the module management module to generate a personalized report template.

6. The method for automatic interpretation and management of tumor gene detection reports according to claim 4, characterized in that: In the step 2), based on the personalized report template, the upstream data and medical interpretation information of the patient's test sample are obtained by using the data management unit, and an automated interpretation report is generated according to the preset data processing and interpretation logic, and the report data is stored, including the following steps: 2.1) Based on the basic information and test information of the patient's test samples in the personalized report template, retrieve the pathological diagnosis and test results from the upstream data; 2.2) Retrieve the corresponding medical interpretation information based on the pathological diagnosis and test results, and process and interpret the upstream data and medical interpretation information according to the preset data processing and interpretation logic to generate an automated interpretation report.

7. A method for automatic interpretation and management of tumor gene detection reports according to claim 6, characterized in that: In step 2.2), based on the pathological diagnosis and test results, the corresponding medical interpretation information is retrieved, and the upstream data and medical interpretation information are processed and interpreted according to the preset data processing and interpretation logic to generate an automated interpretation report, including: 2.2.1) Using pathological diagnosis and test results as input, retrieve corresponding medical interpretation information according to the preset retrieval logic classification; Among them, the preset calling logic is: ① For target drug matching data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input, and the corresponding targeted drug guidance information is matched in the "target drug matching" category of the medical interpretation information and retrieved into the system, including targeted drugs, drug evidence, evidence type, and sensitivity / resistance; ② For classification / prognosis data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input parameters, and the classification / prognosis analysis information is matched in the "classification / prognosis" category of the medical interpretation information and retrieved into the system, including the prognosis assessment results, evidence, and evidence type; ③ For mutation analysis data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input parameters, and the mutation analysis description information is matched in the "Mutation Analysis" category of the medical interpretation information and retrieved into the system, including the biological function and carcinogenicity of the mutation; ④ For immunotherapy data, the cancer types and gene mutations, TMB, and MSI corresponding to the patient test samples in the upstream data are used as input parameters, and the immunotherapy efficacy-related analytical information is matched in the "immunotherapy" category of the medical interpretation information and retrieved into the system, including whether the gene mutation has a positive or negative effect on immunotherapy; ⑤ For chemotherapy drug guidance data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input, and the chemotherapy drug guidance information is matched from the "chemotherapy drugs" category of the medical interpretation information and retrieved into the system, including the toxic and side effects, effectiveness and medication tips of chemotherapy drugs; ⑥ For clinical trial data, the cancer types and gene mutations corresponding to the patient test samples in the upstream data are used as input parameters, and the clinical trial data with completely consistent mutations, cancer types and cancer subtypes are matched in the "clinical trial" category of the medical interpretation information and retrieved into the system; ⑦ For genetic risk analysis data, the cancer types and gene germline variations corresponding to the patient test samples in the upstream data are used as input parameters, and the drugs, drug use evidence, genetic risk analysis, evidence type, and sensitive / resistant information are matched in the "Genetic Analysis" category of the medical interpretation information and retrieved into the system; 2.2.2) Process the retrieved medical interpretation information and upstream data according to the preset data processing and interpretation logic, and generate an automated interpretation report according to the personalized report template configuration requirements; Among them, the generation of automated interpretation reports includes: ① The retrieved data on typing / prognosis, mutation analysis, immunotherapy, chemotherapy drugs, and genetic risk analysis are automatically filled into the corresponding submodules of the personalized report template after processing; ② For target drug matching data and clinical trial data, they are processed according to the preset data processing and interpretation logic and then filled into the targeted drug guidance submodule and clinical trial submodule respectively; Among them, when processing target drug matching data, it includes: Determine the evidence level and variation type of target drug information and related evidence in the medical interpretation information based on preset report display rules and automated judgment logic; Screen target drugs and evidence according to preset automated logic rules for screening and sorting target drugs and evidence, as well as the determined evidence level and variation type of each piece of evidence; The preset automated logic rules for drug targets and evidence screening and sorting include: a. According to the sensitivity or resistance of drugs and evidence, they are automatically displayed in the report in two sections, with all sensitive drugs and evidence in the front and resistant drugs and evidence in the back; b. For the same drug, when the medical interpretation information has high-level evidence, the lower-level evidence will not be output; c. Evidence for the same drug is presented together; d. Sort different evidence by level; e. Evidence of the same level is sorted by variant type; clinical trial phases are arranged from high to low; f. When encountering contradictory evidence for the same drug, the evidence for target drug analysis shall be ranked according to the level of the evidence itself; The screened target drugs and evidence are further screened and sorted according to the preset automated logic for judging, grading and displaying contradictory evidence.

Citation Information

Cited By

  • Pathological gene detection report generation system based on natural language processing

    CN119993368A