Lung Segmentation on Standard and Mobile Chest Radiographs using Oriented Gaussian Derivatives Filter

Abstract

Unsupervised lung segmentation method is one of the mandatory processes in order to develop a Content Based Medical Image Retrieval System (CBMIRS) of CXR. There is limited study found on segmentation of mobile chest radiographs, that is relatively important especially for very sick patients whenever their radiographs will be taken using portable X-Ray machine.The purpose of the study is to present a solution for lung segmentation of standard and mobile chest radiographs using fully automated unsupervised method, based on oriented Gaussian derivatives filter with seven orientations, combined with Fuzzy C-Means clustering and thresholding to refine the lung region. A new algorithm to automatically generate a threshold value for each Gaussian response is also proposed. The algorithms are applied to both PA and AP chest radiographs from both public JSRT and private datasets from collaborative hospital. Two pre-processing blocks are introduced to standardize the images from different machines. Comparisons with the previous works found in the literature on JSRT dataset shows that our method gives a reasonably good result. Performance measures (accuracy, F-score, precision, sensitivity and specificity) for the segmentation of lung in public JSRT dataset are above 0.90 except for the overlap measure is 0.87. The median of overlap score for the private image database is 0.83 (standard machine) and 0.75 (mobile machines). The algorithm is also fast, with the average execution time of 12.5s. Our proposed method is fully automated, unsupervised, with no training or learning stage is necessary to segment the lungs taken using both a standard and mobile machines, and useful for the application of the CBMIRS.