A brain CTP image structured diagnosis atlas generation method and device
By combining a time-parameter convolutional neural network and the medical rule function R(x)=α·CBF+β·MTT, a structured diagnostic atlas and report of brain CTP images are generated, which solves the problems of insufficient consideration of individual differences and large result errors in the existing technology, and realizes efficient and reliable diagnosis of cerebrovascular diseases.
CN122201727APending Publication Date: 2026-06-12BEIJING TIANTAN HOSPITAL AFFILIATED TO CAPITAL MEDICAL UNIV
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- BEIJING TIANTAN HOSPITAL AFFILIATED TO CAPITAL MEDICAL UNIV
- Filing Date
- 2026-03-17
- Publication Date
- 2026-06-12
Smart Images
Figure CN122201727A_ABST
Abstract
The application provides a brain CTP image structured diagnosis atlas generation method and device. The method extracts the time sequence characteristics and perfusion parameter characteristics of a brain CTP sequence image through a time-parameter joint convolutional neural network, generates a lesion probability map, filters a lesion candidate area by using a medical rule function R(x)=a.CBF+beta.MTT on the probability map, and generates a structured diagnosis atlas and report containing a lesion area labeled image and a text description by an atlas output module, which can be output in DICOM, PDF or JSON format. The whole process is scheduled by a unified control module to sequentially operate each functional module, so that automatic identification of a brain ischemic lesion and structured result output are realized.
Need to check novelty before this filing date? Find Prior Art