A gene detection report analysis method and system based on multi-template adaptation

By employing a multi-template adaptive gene testing report parsing method, which combines template matching, OCR/NLP parsing, and manual review, the scalability, accuracy, and robustness issues of automatic gene testing report parsing are resolved, achieving efficient and reliable gene testing report parsing and data output.

CN122417271APending Publication Date: 2026-07-17CHONGQING UNIV CANCER HOSPITAL
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
CHONGQING UNIV CANCER HOSPITAL
Filing Date
2026-06-08
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing automatic gene testing report parsing technologies suffer from poor scalability, low recognition accuracy, inability to guarantee compliance of key field formats, lack of quantitative evaluation mechanisms for the reliability of parsing results, and insufficient robustness.

Method used

A gene testing report parsing method based on multi-template adaptation is adopted. Through template matching, general OCR/NLP parsing, rule verification and manual review, combined with a multi-dimensional confidence scoring mechanism, automatic parsing and reliability assurance of multi-source heterogeneous reports are achieved.

Benefits of technology

It improves the coverage and accuracy of automatic parsing of multi-source heterogeneous gene detection reports, ensures the reliability and format compliance of key medical data, has continuous optimization capabilities, and reduces long-term usage costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122417271A_ABST
    Figure CN122417271A_ABST
Patent Text Reader

Abstract

The application provides a gene detection report analysis method and system based on multi-template adaptation, and relates to the technical field of medical information processing. The method comprises the following steps: obtaining a gene detection report file and performing file format identification and preprocessing; performing feature matching between the visual features, text features, layout features and metadata features of the report and target templates in a preset template library; when the matching is successful, extracting key fields of the gene detection according to the field mapping rules predefined by the target templates; when the matching fails, entering a general analysis channel for analysis; performing multi-dimensional confidence score on the extracted and analyzed results according to the field completeness, field format compliance and field consistency; and automatically routing the analysis results to a processing channel for storage and output in a standardized structured data format according to the multi-dimensional confidence score. The application can improve the automatic analysis coverage, field format compliance, result traceability and reliability of downstream medical data processing of multi-source heterogeneous gene detection reports.
Need to check novelty before this filing date? Find Prior Art