Extraction of sputum cells using thresholding techniques for lung cancer detection

This paper deals with an improved version of a filtering thresholding algorithm for extracting the sputum cell from the raw sputum image for lung cancer early detection. In this method the problem is viewed as a segmentation problem focusing on the extraction of such sputum cells from the images. This will be done by segmenting the image into sputum cell region which includes the nuclei, cytoplasm and the background that includes all the rest. These cells can then be analyzed to see if they are cancerous or not. A database with hundred of sputum color images from normal and abnormal subjects that prepared by the Papanicalaou standard staining method were used. These images were used in testing the new algorithm and the results were compared with the ground truth image of extracted sputum cells. It was noted that the new algorithm offered better extraction results than the previous work. It succeeded in extracting the nuclei and cytoplasm regions. Moreover, it succeeded in determining the best range of thresholding values.