IBSR 2018 Brain Tissue Segmentation

In this project, we employed two segmentation approaches for three main brain tissues—cerebrospinal fluid (CSF), gray matter (GM), and white matter (WM)—using the IBSR 18 dataset. The first approach involved a multi-atlas technique, while the second utilized deep learning, specifically the nnUnet neural network. Following extensive experiments, the deep learning approach demonstrated superior performance with Dice scores of 0.92 for CSF, 0.88 for GM, and 0.94 for WM. These impressive results highlight the effectiveness of the nnUnet neural network in accurately segmenting brain tissues.

You can find the source code in the following Github repository.