A disease risk prediction method and device

By introducing feedback verification units and stacked machine learning models into IPMN disease risk prediction, the problems of insufficient accuracy and interpretability in existing technologies are solved, and highly accurate and reliable disease risk assessment is achieved.

CN122201799APending Publication Date: 2026-06-12THE NAVAL MEDICAL UNIV OF PLA
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
THE NAVAL MEDICAL UNIV OF PLA
Filing Date
2026-05-15
Publication Date
2026-06-12

AI Technical Summary

⚠Technical Problem

In existing technologies, the accuracy and interpretability of disease risk prediction for intraductal papillary myxoma of the pancreas (IPMN) are poor. Current guidelines rely on invasive examinations and are difficult to interpret, resulting in insufficient diagnostic accuracy and low clinical adoption.

⚗Method used

The target feature extraction model includes a feedback verification unit, which verifies the candidate structured feature data. Combined with a stacked machine learning model, predictions are made, and interpretable result explanation data is generated, including feature contribution values, risk level information, and textual explanation information.

🎯Benefits of technology

It improves the accuracy and interpretability of disease risk prediction, reduces the need for invasive examinations, and enhances clinicians' trust and the credibility of the prediction model.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122201799A_ABST
    Figure CN122201799A_ABST
Patent Text Reader

Abstract

The embodiment of the application provides a disease risk prediction method and device, which are applied to a pancreatic ductal intraductal papillary mucinous neoplasm malignant transformation risk prediction process. The method comprises the following steps: obtaining clinical data corresponding to a target patient; inputting the clinical data into a target feature extraction model to obtain candidate structured feature data; wherein the target feature extraction model comprises a feedback verification unit; performing data verification on the candidate structured feature data based on the feedback verification unit to obtain target structured data; inputting the target structured data into a target prediction model to obtain a disease prediction probability; and performing interpretive analysis based on the target structured data and the disease prediction probability to generate result interpretation data corresponding to the clinical data. Through the above technical features, the accuracy of disease risk prediction can be improved while the interpretability of disease risk prediction is improved.
Need to check novelty before this filing date? Find Prior Art