A body organ fdg-pet image segmentation system
By constructing anatomical feature bundles and metabolic fine structure flow, and combining dynamic equilibrium parameters and multi-scale prediction fusion, the boundary blurring and sample difference problems of the body organ FDG-PET image segmentation system are solved, and high-precision organ segmentation is achieved.
CN122089763BActive Publication Date: 2026-07-24HANGZHOU PANORAMIC MEDICAL IMAGING DIAGNOSIS CO LTD
View PDF 2 Cites 0 Cited by
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HANGZHOU PANORAMIC MEDICAL IMAGING DIAGNOSIS CO LTD
- Filing Date
- 2026-04-23
- Publication Date
- 2026-07-24
Smart Images
Figure CN122089763B_ABST
Abstract
The application discloses a kind of body organ FDG-PET image segmentation systems, including image acquisition module, encoder feature extraction module, multi-feature beam decoding collection module, scale prediction fusion module, loss function design module, PET image segmentation model design module and body organ FDG-PET image segmentation module.The application belongs to the field of image processing, specifically refers to a kind of body organ FDG-PET image segmentation system, the scheme is ensured that the segmentation result will not deviate from the overall profile of organ by constructing anatomical structure feature beam;Metabolic fine structure flow is described metabolic activity spatial distribution details;Through hierarchical multi-scale prediction fusion mechanism, avoid the direct pollution of scale mutation;Introduce dynamic balance parameter, establish boundary profile accuracy loss, ensure that the segmentation boundary is strongly related with the real metabolic activity change gradient in image;Further improve body organ FDG-PET image segmentation effect.
Need to check novelty before this filing date? Find Prior Art