Automated Detection of Pneumothorax Using Frontal Chest X-rays

Abstract

Pneumothorax detection can assist doctors in treatment and diagnosis of diseases more accurately. In this paper automated method employed image segmentation techniques for the detection. The preprocessing methods are handled by image processing techniques using MATLAB 2020b software Support Vector machine is applied to classify normal and abnormal lung chest X-ray. Features are extracted from lung image with the texture based segmentation techniques. The rib boundaries are identified with sobel edge detection. The gray level incurrence matrices segmentation method increase accuracy rate with AUC 0.91.