Dilated Inception U-Net (DIU-Net) for Brain Tumor Segmentation
This work addresses the challenging problem of accurate brain tumor segmentation for medical diagnosis and treatment planning, representing an incremental improvement over existing methods.
The authors tackled brain tumor segmentation in MRI by proposing a U-Net-based architecture with Inception modules and dilated convolutions, achieving significantly better performance than state-of-the-art U-Net models for tumor core and whole tumor segmentation on the BraTS 2018 dataset.
Magnetic resonance imaging (MRI) is routinely used for brain tumor diagnosis, treatment planning, and post-treatment surveillance. Recently, various models based on deep neural networks have been proposed for the pixel-level segmentation of tumors in brain MRIs. However, the structural variations, spatial dissimilarities, and intensity inhomogeneity in MRIs make segmentation a challenging task. We propose a new end-to-end brain tumor segmentation architecture based on U-Net that integrates Inception modules and dilated convolutions into its contracting and expanding paths. This allows us to extract local structural as well as global contextual information. We performed segmentation of glioma sub-regions, including tumor core, enhancing tumor, and whole tumor using Brain Tumor Segmentation (BraTS) 2018 dataset. Our proposed model performed significantly better than the state-of-the-art U-Net-based model ($p<0.05$) for tumor core and whole tumor segmentation.