Sputum image detection and extraction for lung cancer early diagnosis

In this paper, we address the problem of detection and extraction sputum cells that help in lung cancer early diagnosis using respectively, a thresholding technique and a Bayesian classification. In the proposed methods the problem is viewed as a segmentation problem focus on extracting such sputum cells from the images whereby we want to partition the image into sputum cell region includes the nuclei, cytoplasm and the background that includes all the rest. These cells will be analyzed to check whether they are cancerous or not. In this study, we used a database of 100 sputum color images to test the proposed methods by comparing it with the ground truth data of extracted sputum cells. Thus a Bayesian classifier has shown a better extraction results, it outperforms the thresholding classifier by allowing a systematic setting of the classification parameter. We analyzed the performance of these methods with respect to the color space representation. We used some performance criteria such as sensitivity, specificity and accuracy to evaluate the proposed methods. Experiments show that performance accuracy of the Bayesian classifier reaches 99% for the sputum cell extraction.