CN-122025203-A - Clinical test data quality control method, device, equipment and medium

CN122025203ACN 122025203 ACN122025203 ACN 122025203ACN-122025203-A

Abstract

The application discloses a quality control method, a device, equipment and a medium for clinical test data, which relate to the field of clinical test data management, and the method comprises the steps of creating a quality control checking rule database and a quality control problem class library; the quality control checking rule database comprises a plurality of minimum granularity rules, the quality control problem category library comprises a plurality of quality control problem categories, the quality control rule library is established according to the minimum granularity rules in the quality control checking rule database and the quality control problem categories in the quality control problem category library and preset checking requirements, diagnosis and treatment data of clinical test subjects are acquired, checking and matching processing is carried out on the diagnosis and treatment data through the quality control rule library to obtain quality control records, and the quality control records are sent to a target user side so that the target user side can carry out checking processing according to the quality control records to generate a quality control report of the diagnosis and treatment data. The application saves time, reduces cost and further improves the quality control efficiency of clinical test data.

Inventors

PENG SUI
TANG LEI

Assignees

中山大学附属第一医院

Dates

Publication Date: 20260512
Application Date: 20260116

Claims (10)

1. A quality control method for clinical trial data, the quality control method comprising: creating a quality control checking rule database and a quality control problem class library, wherein the quality control checking rule database comprises a plurality of minimum granularity rules; Creating a quality control rule base according to a minimum granularity rule in the quality control checking rule database and a quality control problem category in the quality control problem category base and a preset checking requirement; the diagnosis and treatment data of the clinical test subjects are obtained, wherein the diagnosis and treatment data comprise at least one of medical records, test reports and medication records; Checking and matching treatment is carried out on the diagnosis and treatment data through the quality control rule base, so that a quality control record is obtained; And sending the quality control record to a target user side, so that the target user side performs checking processing according to the quality control record, and generates a quality control report of the diagnosis and treatment data.
2. The method of claim 1, wherein creating a quality control verification rule database comprises: acquiring a quality control checking rule list; converting the quality control checking rule list into a plurality of structured rule entries, wherein the structured rule entries comprise rule descriptions, rule types, rule formats, rule items, change attributes, change times and time thresholds; and combining and sorting single or multiple structured rule entries into a minimum granularity rule to create a quality control checking rule database.
3. The method of claim 1, wherein creating a quality control problem category library comprises: Acquiring a clinical trial quality control problem scene, a system flow engine event and a trial participant role, wherein the system flow engine event comprises a problem processing flow and a problem feedback flow, and the trial participant role comprises a main researcher, a clinical inspector and a mechanism quality controller; And configuring a corresponding problem early warning mode, a problem quality control level and a problem category name according to the clinical trial quality control problem scene, the system flow engine event and the trial participant roles so as to create the quality control problem category library.
4. The method of claim 1, wherein creating a quality control rule base according to a preset verification requirement according to a minimum granularity rule in the quality control verification rule database and a quality control problem category in the quality control problem category base, comprises: Combining single or multiple minimum granularity rules in the checking rule database according to checking requirements to form a checking logic item; Configuring corresponding quality control rule attributes for the checking logic items based on quality control problem categories in the quality control problem category library, wherein the quality control rule attributes comprise rule names, the category of the problems, problem risk levels, applicable period categories and using quality control links; And creating the quality control rule base based on the quality control rule attribute and the check logic item.
5. The method according to claim 3, wherein the quality control record comprises a target quality control problem and detailed problem information; And checking and matching the diagnosis and treatment data through the quality control rule base to obtain a quality control record, wherein the quality control record comprises the following steps: Performing information extraction on the diagnosis and treatment data by adopting a trained natural language processing model to obtain extracted contents, wherein the information extraction comprises entity identification, relation extraction and attribute extraction; Normalizing, converting, deriving and organizing the extracted content to obtain identifiable structured data; based on the structured data, carrying out piece-by-piece checking and matching processing through the quality control rule base, and judging whether the matching is successful or not; If the matching is successful, marking the category of the question which is successfully matched as a target quality control question, and recording the detailed information of the question of the target quality control question, wherein the detailed information of the question comprises the quality control question category, the occurrence position of the question, diagnosis and treatment data and a judgment basis.
6. The method for quality control of clinical trial data according to claim 5, wherein the target user side comprises a first user side and a second user side, the first user side is a device side corresponding to a main researcher, and the second user side is a device side corresponding to a facility quality inspector; the quality control record is sent to a target user side, so that the target user side performs checking processing according to the quality control record, and a quality control report of the diagnosis and treatment data is generated, and the quality control report comprises the following steps: determining the problem processing flow and the receiving roles of the flows associated with the quality control problem category; pushing the target quality control problem to a first user side corresponding to the current receiving role according to the problem processing flow, enabling the first user side to check corresponding problem detailed information according to the target quality control problem, calling traceability data to locate problem data based on the problem occurrence position, judging the accuracy of the target quality control problem, generating a quality control problem list and sending the quality control problem list to a second user side; and receiving a quality control report of the diagnosis and treatment data through the first user side, wherein the quality control report is obtained by checking and sorting the quality control problem list according to a correction feedback flow associated with the quality control problem category by the second user side.
7. The method for quality control of clinical trial data according to claim 6, characterized in that the method further comprises: And setting a rectification plan and correcting the quality control problem through the first user side according to the quality control report, generating a rectification report and sending the rectification report to the second user side, so that the second user side checks the quality control problem according to the rectification report, confirms whether to rectify the quality control problem, and generates a rectification report.
8. A clinical trial data quality control device, characterized in that the clinical trial data quality control device comprises: The system comprises a first creating module, a quality control checking rule database and a quality control problem category database, wherein the quality control checking rule database comprises a plurality of minimum granularity rules; The second creating module is used for constructing a quality control rule base according to the quality control checking rule database and the quality control problem category base and preset checking requirements; The system comprises an acquisition module, a diagnosis and treatment module and a diagnosis and treatment module, wherein the acquisition module is used for acquiring diagnosis and treatment data of a clinical test subject, and the diagnosis and treatment data comprises at least one of medical records, test reports and medication records; the checking processing module is used for checking and matching the diagnosis and treatment data through the quality control rule base to obtain a quality control record; and the quality control report generating module is used for sending the quality control record to a target user side so that the target user side performs checking processing according to the quality control record to generate a quality control report of the diagnosis and treatment data.
9. A computer device comprising a memory, a processor to store a computer program executable on the memory, wherein the processor executes the computer program to perform the steps of the clinical trial data quality control method of any one of claims 1-7.
10. A computer readable storage medium having stored thereon a computer program, which, when executed by a processor, implements the clinical trial data quality control method of any one of claims 1-7.

Description

Clinical test data quality control method, device, equipment and medium Technical Field The application relates to the technical field of clinical test data management, in particular to a quality control method, device, equipment and medium of clinical test data. Background With the continuous development of medical technology and the increase of clinical trial complexity, a great deal of clinical trial data is generated in the verification process of the effectiveness and safety of drugs or therapies. The quality of clinical test data directly affects success and failure of drug development and safety of patients, and quality control of the clinical test data is a key link for ensuring research accuracy and reliability of drug development, medical equipment and the like, and how to control the quality of the clinical test data is particularly important for ensuring the reliability of the data. At present, in the related art, manual experience is adopted to control the quality of clinical test data, but partial outpatient subjects are easy to deviate when self-recording medicines or symptoms, so that the data is insufficient in authenticity, difficult to trace, high in recording error rate, large in data cannot be covered on the whole by adopting the scheme, the subjectivity is high depending on the manual experience, time and cost are consumed, and further the quality control efficiency is low. Disclosure of Invention The application aims to provide a quality control method, device, equipment and medium for clinical test data. In order to achieve the above object, the present application provides the following solutions: In a first aspect, the present application provides a method for quality control of clinical trial data, comprising: creating a quality control checking rule database and a quality control problem class library, wherein the quality control checking rule database comprises a plurality of minimum granularity rules; Creating a quality control rule base according to a minimum granularity rule in the quality control checking rule database and a quality control problem category in the quality control problem category base and a preset checking requirement; the diagnosis and treatment data of the clinical test subjects are obtained, wherein the diagnosis and treatment data comprise at least one of medical records, test reports and medication records; Checking and matching treatment is carried out on the diagnosis and treatment data through the quality control rule base, so that a quality control record is obtained; And sending the quality control record to a target user side, so that the target user side performs checking processing according to the quality control record, and generates a quality control report of the diagnosis and treatment data. In one embodiment, creating a quality control verification rule database includes: acquiring a quality control checking rule list; converting the quality control checking rule list into a plurality of structured rule entries, wherein the structured rule entries comprise rule descriptions, rule types, rule formats, rule items, change attributes, change times and time thresholds; and combining and sorting single or multiple structured rule entries into a minimum granularity rule to create a quality control checking rule database. In one embodiment, creating a quality control problem category library includes: Acquiring a clinical trial quality control problem scene, a system flow engine event and a trial participant role, wherein the system flow engine event comprises a problem processing flow and a problem feedback flow, and the trial participant role comprises a main researcher, a clinical inspector and a mechanism quality controller; And configuring a corresponding problem early warning mode, a problem quality control level and a problem category name according to the clinical trial quality control problem scene, the system flow engine event and the trial participant roles so as to create the quality control problem category library. In one embodiment, creating the quality control rule base according to the minimum granularity rule in the quality control checking rule database and the quality control problem category in the quality control problem category base and the preset checking requirement includes: Combining single or multiple minimum granularity rules in the checking rule database according to checking requirements to form a checking logic item; Configuring corresponding quality control rule attributes for the checking logic items based on quality control problem categories in the quality control problem category library, wherein the quality control rule attributes comprise rule names, the category of the problems, problem risk levels, applicable period categories and using quality control links; And creating the quality control rule base based on the quality control rule attribute and the check logic item. In one embodiment, the quality contr