Enhancing MRI Brain Tumor Segmentation with an Additional Classification Network

Hieu T. Nguyen, Tung T. Le, Thang V. Nguyen, Nhan T. Nguyen

Abstract: Brain tumor segmentation plays an essential role in medical image analysis. In recent studies, deep convolution neural networks (DCNNs) are extremely powerful to tackle tumor segmentation tasks. We propose in this paper a novel training method that enhances the segmentation results by adding an additional classification branch to the network. The whole network was trained end-to-end on the Multimodal Brain Tumor Segmentation Challenge (BraTS) 2020 training dataset. On the BraTS’s test set, it achieved an average Dice score of $$80.57\%$$ , $$85.67\%$$ and $$82.00\%$$ , as well as Hausdorff distances $$(95\%)$$ of 14.22, 7.36 and 23.27, respectively for the enhancing tumor, the whole tumor and the tumor core.