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.
Description:
This thesis submitted in partial fulfillment of the requirements for the degree of Bachelor of Science in Information and Communication Engineering of East West University, Dhaka, Bangladesh
