Biomedical Image Segmentation Using Deep Learning

Abstract

The process of solving medical issues by evaluating images created in clinical workout is known as medical image analysis.The goal is to obtain data for enhanced clinical diagnosis in an effective way. In this article presents a systematic overview of the present advance stage of medical image analysis using deep convolutional networks and also aims to develop automated methods for hippocampus brain MRI segmentation to decrease the time-consuming workload done by radiologist. We compared our results with hand-labeled segmentation done by medical radiologist. Our automated segmentation agreed well with human raters using a leave-one-out approach and standard overlap and distance error metrics,any differences were comparable with differences between trained human raters. Our error metrics compare favorably with those previously reported for other automated segmentations of hippocampus, suggesting the effectiveness of the approach to large-scale studies.