Sequential 3D U-Nets for Biologically-Informed Brain Tumor Segmentation

TL;DR

This paper introduces a biologically-informed 3D U-Net approach for brain tumor segmentation that models glioma tissue structure, resulting in improved accuracy on the BraTS dataset.

Contribution

It presents a novel tree-structured deep neural network architecture that incorporates biological tissue relationships for more accurate glioma segmentation.

Findings

Achieved Dice scores of 0.882 for whole tumor

Achieved Dice scores of 0.732 for enhancing tumor

Achieved Dice scores of 0.730 for tumor core

Abstract

Deep learning has quickly become the weapon of choice for brain lesion segmentation. However, few existing algorithms pre-configure any biological context of their chosen segmentation tissues, and instead rely on the neural network's optimizer to develop such associations de novo. We present a novel method for applying deep neural networks to the problem of glioma tissue segmentation that takes into account the structured nature of gliomas - edematous tissue surrounding mutually-exclusive regions of enhancing and non-enhancing tumor. We trained multiple deep neural networks with a 3D U-Net architecture in a tree structure to create segmentations for edema, non-enhancing tumor, and enhancing tumor regions. Specifically, training was configured such that the whole tumor region including edema was predicted first, and its output segmentation was fed as input into separate models to predict enhancing and non-enhancing tumor. Our method was trained and evaluated on the publicly available BraTS dataset, achieving Dice scores of 0.882, 0.732, and 0.730 for whole tumor, enhancing tumor and tumor core respectively.