Tumor region segmentation method in CT images based on radiomics scoring feature map
CN118314156BActive Publication Date: 2024-12-31THE AFFILIATED HOSPITAL OF SOUTHWEST MEDICAL UNIV
View PDF 0 Cites 0 Cited by
Patent Information
- Application Number
- CN202410555086.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-07
- Publication Date
- 2024-12-31
- Estimated Expiration
- 2044-05-07
AI Technical Summary
Technical Problem
Traditional manual segmentation methods for tumors in CT images rely on clinician experience and subjective judgment, leading to inefficiencies and inaccuracies, while convolutional neural networks (CNN) lack transparency and clinical applicability due to unexplainable feature learning.
Method used
A CT image tumor region segmentation method using radiomics scoring feature maps, involving feature extraction, screening, constructing radiomics scores, and generating masks to automate tumor segmentation, providing a transparent and accurate segmentation strategy.
Benefits of technology
The method enhances segmentation accuracy and provides biological reasoning, improving efficiency and transparency by using radiomics scores to distinguish tumors from surrounding tissues.
✦ Generated by Eureka AI based on patent content.
Smart Images
Figure CN118314156B_ABST
Abstract
The present invention provides a method for segmenting tumor regions in CT images based on an imaging omics scoring feature map, which belongs to the field of image processing technology. The method includes extracting imaging group features of tumors and surrounding tissues in CT images; screening imaging omics features that can distinguish tumors and surrounding tissues based on the extracted imaging group features of tumors and surrounding tissues; constructing an imaging omics score based on the screened imaging omics features; establishing an imaging omics scoring feature map based on the constructed imaging omics score; generating a mask based on the established imaging omics scoring feature map, and automatically segmenting the tumor region of the CT image based on the generated mask. The present invention solves the problem that traditional manual segmentation methods often rely on the experience and subjective judgment of clinicians and have limitations in efficiency and accuracy.
Need to check novelty before this filing date? Find Prior Art