Method and system for predicting postoperative endocrine function deficiency of sellar region tumor

By using multi-source data fusion and multi-label classification networks, the problems of narrow pathological type coverage and single prediction axis in the prediction of postoperative endocrine dysfunction in sellar region tumors were solved. This enabled accurate prediction and early intervention of multi-axis endocrine function, reducing the safety risks of postoperative endocrine deficits.

CN122417360APending Publication Date: 2026-07-17BEIJING CHILDRENS HOSPITAL AFFILIATED TO CAPITAL MEDICAL UNIV
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
BEIJING CHILDRENS HOSPITAL AFFILIATED TO CAPITAL MEDICAL UNIV
Filing Date
2026-04-21
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing technologies for predicting endocrine dysfunction after surgery for sellar region tumors suffer from problems such as narrow coverage of pathological types, single prediction axis, lack of intraoperative parameter fusion, and inability to provide clinical decision support, resulting in cumbersome provocation test procedures and serious safety risks.

Method used

By acquiring and preprocessing multi-source data, extracting image features, encoding pathological type conditional parameters, and predicting multi-label endocrine deficits, combined with risk stratification and clinical decision support, a system for predicting postoperative endocrine function deficits in sellar region tumors was constructed. This system can predict the probability of multi-axis endocrine axis deficits in various sellar region tumor types and provide interpretable reports.

Benefits of technology

It has achieved a unified prediction framework for major sellar region tumor types such as pituitary adenoma and craniopharyngioma, which significantly improves prediction accuracy. It can output a report on the probability of defects in five endocrine axes on the first day after surgery, guide early replacement therapy, and reduce the risk of serious complications.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122417360A_ABST
    Figure CN122417360A_ABST
Patent Text Reader

Abstract

本发明公开了鞍区肿瘤术后内分泌功能缺损预测方法及其系统，涉及医学人工智能技术领域，该方法包括：多源数据采集与预处理，获取术前MRI影像、临床病理和术中操作三组参数；影像特征提取，自动分割肿瘤并提取垂体柄空间关系等多维影像特征；病理类型条件化参数编码，根据肿瘤病理类型激活差异化编码路径；多标签内分泌缺损预测，通过五个独立sigmoid输出节点同时预测五轴缺损概率；风险分层与临床决策支持，根据差异化阈值输出分层干预建议和SHAP可解释性报告。
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Early evaluation method for endocrine remission of pituitary growth hormone adenoma based on low Knosp grading

    CN121034617A