Pneumothorax Detection in Chest Radiographs Using Local and Global Texture Signatures

 

Overview

Pneumothorax is one of the most common thoracic injuries and poses a risk of serious morbidity. It is manifested as an abnormal accumulation of air in the pleural space that separates the lung from the chest wall. Symptoms range from asymptomatic to life-threatening respiratory distress. A novel framework for automatic detection of pneumothorax abnormality on chest radiographs is presented. The suggested method is based on a texture analysis approach combined with supervised learning techniques. The proposed framework consists of two main steps: at first, a texture analysis process is performed for detection of local abnormalities. Labeled image patches are extracted in the texture analysis procedure following which local analysis values are incorporated into a novel global image representation. The global representation is used for training and detection of the abnormality at the image level. The presented global representation is designed based on the distinctive shape of the lung, taking into account the characteristics of typical pneumothorax abnormalities. A supervised learning process was performed on both the local and global data, leading to trained detection system. Several state of the art texture feature sets were experimented with (Local Binary Patterns, Maximum Response filters). The performance was measured by receiver operating characteristic (ROC) curves, obtaining area under curve values of 0.88 and 0.82 for right and left pneumothorax respectively.

 

 

Publications

Geva, O., Zimmerman-Moreno, G., Lieberman, S., Konen, E., & Greenspan, H. (2015, March). Pneumothorax detection in chest radiographs using local and global

texture signatures. In SPIE Medical Imaging (pp. 94141P-94141P). International Society for Optics and Photonics.

horizontal rule

For problems or questions regarding this web page contact hayit@eng.tau.ac.il.
Last updated: 19/11/15.