Segmentation of Large Scale Medical Images using HPC: Classification of Methods and Challenges

Medical imagining is one of the major disciplines for analyzing human tissues non-invasively. Image segmentation is a sub-process in image processing that divides the given image into meaningful regions that can used for further classification and analysis. This step is challenging, as it difficult to identify precisely and extract that portion of the image having abnormal tissues for further diagnosis and analysis. There are several methods and techniques that exist for image segmentation. But not all algorithms and techniques can be applied on medical images. Owing to the growth of medical image corpus and the need for automated image techniques, there is a need for large scale image segmentation techniques that can precisely identify the region-of-interest in real-time for diagnosis. This paper presents a comprehensive survey and review of the medical image segmentation models, techniques, algorithms and challenges that exists from medium and large scale image processing perspectives. We have explored the large scale segmentation using parallel and distributed computing platforms for handling data with ease with reduced cost. Keywords— Medical image analysis, Medical image segmentation, Large scale image processing.