Related Experiment Video
Updated: Dec 14, 2025
07:53
Author Spotlight: Advancing 3D Modeling for Enhanced Diagnosis and Treatment of Pulmonary Nodules in Early-Stage Lung Cancer
Published on: October 13, 2023
1.9K
Toward an Affine Feature-Based Registration Method for Ground Glass Lung Nodule Tracking
Yehuda Kfir Ben Zikri1, María Helguera1, Nathan D Cahill2
1Chester F. Carlson Center for Imaging Science, Rochester Institute of Technology, Rochester, NY, USA.
Vipimage 2019 : Proceedings of the VII ECCOMAS Thematic Conference on Computational Vision and Medical Image Processing, October 16-18, 2019, Porto, Portugal. Vipimage (Conference) (2019 : Porto, Portugal)
|July 24, 2020
Summary
This study introduces a new method for tracking lung nodule changes using registered CT scans. The technique improves accuracy in assessing tumor progression or regression, aiding clinical decisions.
Area of Science:
- Medical Imaging
- Radiology
- Computational Biology
Background:
- Accurate lung nodule progression assessment is vital for disease management and therapy response evaluation.
- Computed tomography (CT) is standard for lung nodule tracking, but segmentation errors due to irregular boundaries complicate quantification.
- Existing methods struggle with background lung tissue deformation, impacting nodule assessment accuracy.
Purpose of the Study:
- To develop and evaluate a feature-based affine image registration framework for accurate lung nodule progression assessment.
- To improve the quantification of lung nodule changes by accounting for thoracic CT image deformation.
- To enhance clinical confidence in evaluating tumor progression or regression.
Main Methods:
- Developed a feature-based affine image registration framework to align serial thoracic CT images.
- Utilized digital subtraction images post-registration to assess nodule changes.
- Evaluated the method on twelve de-identified patient datasets.
Main Results:
- Achieved registration accuracy better than 1.5mm compared to non-rigid registration techniques.
- Demonstrated the framework's ability to account for background lung tissue deformation.
- Clinical assessment using registered subtraction images showed consistency and increased confidence over visual analysis.
Conclusions:
- The proposed feature-based affine registration framework offers a robust and accurate method for lung nodule progression assessment.
- This technique enhances the reliability of quantifying tumor changes from thoracic CT images.
- The method shows significant clinical potential for improving nodule tracking and patient management.
