Rethinking Radiology: An Analysis of Different Approaches to BraTS

TL;DR

This paper compares various deep learning models, including UNet, for glioma segmentation on the BraTS dataset, analyzing their performance and potential improvements with recent architectural advancements.

Contribution

It provides a comprehensive comparison of existing deep learning architectures for glioma segmentation and discusses potential enhancements using recent deep learning innovations.

Findings

UNet performs competitively on BraTS dataset

Certain architectural modifications improve segmentation accuracy

Discussion on future directions for deep learning in radiology

Abstract

This paper discusses the deep learning architectures currently used for pixel-wise segmentation of primary and secondary glioblastomas and low-grade gliomas. We implement various models such as the popular UNet architecture and compare the performance of these implementations on the BRATS dataset. This paper will explore the different approaches and combinations, offering an in depth discussion of how they perform and how we may improve upon them using more recent advancements in deep learning architectures.