Related Experiment Video
Updated: May 25, 2026
10:44
Three-Dimensional Phase Resolved Functional Lung Magnetic Resonance Imaging
Published on: June 21, 2024
Temporal segmentation of lung region from MRI sequences using multiple active contours.
Renato Seiji Tavares1, José Miguel Manzanares Chirinos, Leonardo Ishida Abe
1Computational Geometry Laboratory, Escola Politécnica, São Paulo University, Brazil.
Summary
This study introduces a novel lung segmentation method using active contours and respiratory patterns, improving accuracy despite varying image quality. The technique shows promise for analyzing lung conditions in both healthy individuals and COPD patients.
Area of Science:
- Medical imaging
- Image processing
- Pulmonology
Background:
- Lung segmentation is challenging due to significant variations in medical image quality.
- Accurate lung segmentation is crucial for diagnosing and monitoring respiratory diseases.
Purpose of the Study:
- To develop a robust lung segmentation algorithm that overcomes image quality limitations.
- To accurately capture synchronous respiratory patterns for improved segmentation.
- To validate the algorithm's performance in healthy subjects and Chronic Obstructive Pulmonary Disease (COPD) patients.
Main Methods:
- A modified Hough transform combined with a mask creation algorithm to detect respiratory patterns.
- A greedy active contour algorithm to relax synchronicity constraints.
- Gravitation Vector Field (GVF) active contour algorithm for initial segmentation based on point clouds.
- A final active contour algorithm for boundary refinement.
Main Results:
- The developed algorithm demonstrates robust lung segmentation, even with variable image quality.
- Synchronous respiratory patterns were effectively determined and utilized for segmentation.
- The method was successfully tested on both healthy individuals and COPD patients.
Conclusions:
- The proposed active contour-based lung segmentation method offers a reliable approach for medical image analysis.
- This technique enhances the ability to segment lungs accurately, aiding in the assessment of respiratory conditions.
- Further validation through temporal registration of coronal and sagittal images supports the algorithm's efficacy.
