用于帕金森病理亚型分型的装置及存储介质
By constructing a causal graph classification scheme based on magnetic resonance imaging data, the problems of accuracy and reliability in the classification of Parkinson's pathological subtypes have been solved, and accurate classification in the prodromal stage of the disease has been achieved, providing scientific support for personalized diagnosis and treatment.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING FRIENDSHIP HOSPITAL CAPITAL MEDICAL UNIV
- Filing Date
- 2025-11-26
- Publication Date
- 2026-07-17
AI Technical Summary
Current Parkinson's pathological subtype classification techniques lack objective quantitative indicators, making it impossible to achieve accurate classification in the prodromal stage of the disease. They are easily influenced by physician experience and cannot distinguish the pathological origin differences between body-priority and brain-priority subtypes, thus limiting the accuracy of classification.
By acquiring magnetic resonance imaging data, extracting the magnetic susceptibility values of key brain regions, constructing a standardized feature matrix, initializing a weighted adjacency matrix, constructing a target optimization function containing a fitting function, a sparse regularization term, and a loop-free constraint penalty term, iteratively optimizing the causal graph, forming a specific causal graph, evaluating the causal effect, and realizing the subtyping of Parkinson's pathological types.
This study achieved objective and quantitative classification of Parkinson's disease pathological subtypes, revealed the pathological transmission mechanism, improved the accuracy and reliability of classification, and provided a scientific basis for early intervention and personalized diagnosis and treatment.
Smart Images
Figure CN121730747B_ABST