Related Experiment Video
Updated: Nov 22, 2025
08:05
Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
Published on: December 19, 2020
14.5K
A robust and efficient framework for tubular structure segmentation in chest CT images.
Summary
This study introduces a novel framework for robust pulmonary tubular structure segmentation in chest CT scans. The automated approach enhances accuracy and efficiency in identifying these critical anatomical features.
Area of Science:
- Medical Imaging
- Radiology
- Computer-Aided Diagnosis
Background:
- Pulmonary tubular structure segmentation in chest CT images is crucial for reducing false positives.
- Accurate segmentation improves the classification performance of lung nodule malignancy levels.
Purpose of the Study:
- To present a robust and efficient framework for segmenting pulmonary tubular structures.
- To enhance the accuracy of lung nodule analysis through improved segmentation.
Main Methods:
- A global tubular structure identification model using the Frangi filter was developed.
- A local tubular structure identification model with a sliding window was designed for vessel segmentation.
- A multi-view voxel discriminating scheme was proposed to reduce computational complexity.
Main Results:
- The framework achieved a True Positive Rate (TPR) of 85.79%.
- A False Positive Rate (FPR) of 24.83% was recorded.
- An Accuracy (ACC) of 84.47% was obtained with an average processing time of 162.9 seconds.
Conclusions:
- The proposed framework offers an automated solution for effective tubular structure segmentation in chest CT.
- This method holds potential for improving diagnostic accuracy in pulmonary imaging.
