Related Experiment Video
Updated: May 12, 2026
08:05
Lung CT Segmentation to Identify Consolidations and Ground Glass Areas for Quantitative Assesment of SARS-CoV Pneumonia
Published on: December 19, 2020
Automatic detection of lesions in lung regions that are segmented using spatial relations
Donia Ben Hassen1, Hassen Taleb
1LARODEC, Higher Institute of Management, University of Tunis, Tunisia. donia_ben_hassen@yahoo.fr
Clinical Imaging
|April 23, 2013
Summary
This study introduces a new method for automatically detecting lesions in chest X-rays. Accurate lung segmentation and feature selection significantly improve diagnostic accuracy in Computer Aided Diagnosis (CAD) systems.
Area of Science:
- Medical Imaging
- Computer-Aided Diagnosis
- Radiology
Background:
- Accurate detection of chest lesions is crucial for timely diagnosis.
- Existing Computer Aided Diagnosis (CAD) systems require robust feature extraction.
- Preprocessing steps like segmentation are vital for improving CAD performance.
Purpose of the Study:
- To develop a novel approach for automatic lesion detection and feature selection in chest radiographs.
- To highlight the importance of accurate lung segmentation as a preprocessing step.
- To identify effective features for describing various chest lesions.
Main Methods:
- A novel segmentation approach based on spatial relationships of lung structures was developed.
- A Computer Aided Diagnosis (CAD) scheme was utilized.
- Forward stepwise selection was employed to identify optimal feature combinations from original and transformed images.
- Skeletal structures were suppressed to enhance lesion visibility.
Main Results:
- The proposed segmentation method improved detection accuracy.
- Suppression of skeletal structures further enhanced diagnostic performance.
- Selected features effectively characterized different types of chest lesions.
- Experimental results demonstrated the efficacy of the combined approach.
Conclusions:
- Accurate lung segmentation is a critical preprocessing step for CAD systems.
- The developed feature selection method efficiently identifies discriminative features for lesion detection.
- This approach shows promise for improving the accuracy and reliability of automated chest radiograph analysis.
