Medical Image Segmentation Techniques and Their Relevance in Contemporary Imaging

Image segmentation is a method where the image is partitioned into multiple segments. Segmentation of the image is typically used for locating objects and boundaries in images and produces a set of segments that include the entire image or a series of contours drawn from the image together. Similarly, Medical Image Segmentation focusses on the automated or semi-automatic method of detecting boundaries within a 2D or 3D image. However, the major problem with segmenting the medical image is the high variability in medical images, to overcome which various segmentation techniques may be applied on these images. This paper presents the two most effective segmentation techniques i.e. fuzzy c means clustering and ant colony optimisation-canny on 3D CT scan of human body parts which further shows a comparative analysis of entropy values of the resultant images, with FCM providing drastic rise in the values. This may be highly useful to the medical practitioners for better understanding and visualisation of the images to provide effective diagnosis of the disease.