Liver ct image reconstruction method under low dose scanning condition
By constructing a low-dose liver CT image reconstruction framework based on local structure perception and information regulation, the problem of balancing noise suppression and detail preservation in low-dose liver CT image reconstruction is solved, achieving high-quality reconstruction of liver CT images and improving the recognizability and spatial continuity of liver edge contours and structural details.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- THE FIRST MEDICAL CENT CHINESE PLA GENERAL HOSPITAL
- Filing Date
- 2025-12-24
- Publication Date
- 2026-06-09
AI Technical Summary
Existing low-dose liver CT image reconstruction techniques struggle to achieve a good balance between noise suppression and detail preservation, particularly in terms of insufficient enhancement of liver tissue boundaries and structural details. They fail to fully exploit the deep morphological distribution features of the images, resulting in insufficient accuracy in restoring complex structural details and inadequate ability to express the distribution changes of information in spatial structures.
A low-dose liver CT image reconstruction framework with local structure perception, potential morphological analysis and information modulation capabilities is constructed. Through the structure aggregation module, morphological modulation module and information expansion modulation module, the local feature expression of liver edge contour, texture transition area and low contrast area is enhanced, and finally high-quality reconstructed image is generated.
It significantly improves the clarity and structural continuity of liver CT images under low-dose conditions, enhances the structural discernibility of liver tissue edges and overall spatial coherence, and improves the diagnostic reliability of the images.
Smart Images

Figure CN121639866B_ABST
Abstract
Citation Information
Patent Citations
Super-resolution denoising method and device for a low-dose CT image
CN113516586A
Low-dose CT image super-resolution reconstruction method based on double-branch joint learning
CN115953491A
Method for reconstructing low-dose CT images based on redundant information of standard dose images
CN102063728A
Vertebral CT image standardization preprocessing method and system fusing multi-parameter features
CN120976337A