Redundancy Reduction in Semantic Segmentation of 3D Brain Tumor MRIs
This work addresses brain tumor segmentation for medical analysis and treatment planning, but it is incremental as it builds on existing encoder-decoder networks with training modifications and ensembling.
The paper tackled brain tumor segmentation in 3D MRIs by modifying network training to reduce redundancy and using confidence-based ensembling, achieving top scores of 0.8600, 0.8868, and 0.9265 dice for enhanced tumor core, tumor core, and whole tumor on the BraTS 2021 dataset.
Another year of the multimodal brain tumor segmentation challenge (BraTS) 2021 provides an even larger dataset to facilitate collaboration and research of brain tumor segmentation methods, which are necessary for disease analysis and treatment planning. A large dataset size of BraTS 2021 and the advent of modern GPUs provide a better opportunity for deep-learning based approaches to learn tumor representation from the data. In this work, we maintained an encoder-decoder based segmentation network, but focused on a modification of network training process that minimizes redundancy under perturbations. Given a set trained networks, we further introduce a confidence based ensembling techniques to further improve the performance. We evaluated the method on BraTS 2021 validation board, and achieved 0.8600, 0.8868 and 0.9265 average dice for enhanced tumor core, tumor core and whole tumor, respectively. Our team (NVAUTO) submission was the top performing in terms of ET and TC scores and within top 10 performing teams in terms of WT scores.