Related Experiment Video
Updated: Mar 2, 2026
10:28
Gene Regulation and Targeted Therapy in Gastric Cancer Peritoneal Metastasis: Radiological Findings from Dual Energy CT and PET/CT
Published on: January 22, 2018
11.6K
SU-E-I-87: Tumor Positioning for PET-CT Scanner by Jointly Registration and Segmentation
D Li1,2,3, J Yang1,2,3, Y Yin1,2,3
1College of Physics and Electronics, Shandong Normal University, Ji nan, Shandong Province.
Medical Physics
|May 19, 2017
Summary
An efficient method for tumor positioning in radiotherapy planning uses joint registration and segmentation on 18F-FDG PET-CT scans. This approach accurately transfers tumor contours for improved treatment planning, achieving high overlap with manual segmentations.
Area of Science:
- Medical Imaging
- Radiotherapy
- Computational Anatomy
Background:
- Accurate tumor positioning is crucial for effective radiotherapy planning.
- Current methods may lack efficiency and precision in automated tumor delineation.
- 18F-FDG PET-CT offers functional and anatomical imaging for tumor visualization.
Purpose of the Study:
- To develop an efficient and accurate method for automatic tumor positioning in radiotherapy planning.
- To achieve precise tumor delineation and contour transfer using joint registration and segmentation on 18F-FDG PET-CT scans.
Main Methods:
- A two-stage approach involving region growing segmentation on PET scans and deformable registration to CT scans.
- Utilized SUV monotonous features for segmentation and an edge-preserving scale space (TV-L1) with free-form deformation (FFD) for registration.
- Coarse-to-fine deformation field generation for accurate contour transfer.
Main Results:
- The proposed method achieved segmentation results comparable to manual delineations by radiation oncologists.
- Automatic transfer of gross tumor volume (GTV) contours to CT scans was suitable for radiotherapy planning.
- Average volumetric overlap between automatically and manually segmented tumors ranged from 90% to 97%.
Conclusions:
- The developed joint registration and segmentation method provides an efficient and accurate solution for tumor positioning in 18F-FDG PET-CT datasets.
- This technique holds significant potential for improving radiotherapy planning workflows.
