Computer Based Performance Evaluation of Segmentation Methods for Chest X-Ray Image

Abstract

Abstract-According to WHO (World Health Organization) report, because of chest diseases more than 12 million death cases are reported during the year 2008. If chest disease is detected in its early stage then the possibility of surviving the patient is more. One of the most preferred scan for the early detection of chest disease is X-ray scan. X-ray is inexpensive, painless and required less time to generate image. X-ray image contain a lot of irrelevant information and has intensity problems, which makes the task of locating and analyzing suspicious area difficult by the doctor. By using pre-processing and segmentation technique, suspicious area is easily separate out from the rest of the structure. In the present paper, segmentation techniques and the segmentation results after applying on the X-ray images are discussed. In segmentation, image is partition into a meaningful region. The result of image segmentation is a set of segments that collectively cover the entire image and all pixel in the segmented region which are similar with respect to some characteristic or computed property such as color, intensity, texture etc. In present paper, some of the segmentation techniques such as edge detection, thresholding, skeletonization, contour and watershed transform are applied on the chest X-ray image and the effectiveness of each techniques are shown with the help of images and properties extracted.

CHEST X-RAY SEGMENTATION to CALCULATE PLEURAL EFFUSION INDEX in PATIENT with DENGUE HEMORRHAGIC FEVER

“…The spot determined as the maximum width of right pleural effusion was the upper right diaphragm with the right lung because the effusion fluid will mostly collect in the right costofrenicus corner due to the influence of gravity. PEI assessment can be used as a predictor of the severity of DHF which is a risk factor for mortality [6]. Currently PEI calculation was still done manually by radiologist specialists.…”