Brain Tumor Segmentation using 3D-CNNs with Uncertainty Estimation
This work addresses the problem of high memory consumption and lack of uncertainty information in 3D-CNNs for brain tumor segmentation, which is critical for neurologists in medical diagnosis, but it is incremental as it builds on existing V-Net architectures.
The paper tackles brain tumor segmentation in 3D MRIs by proposing a 3D encoder-decoder architecture with patching to reduce memory consumption and introducing voxel-wise uncertainty estimation to aid medical diagnosis, achieving improved results in segmentation tasks.
Automation of brain tumors in 3D magnetic resonance images (MRIs) is key to assess the diagnostic and treatment of the disease. In recent years, convolutional neural networks (CNNs) have shown improved results in the task. However, high memory consumption is still a problem in 3D-CNNs. Moreover, most methods do not include uncertainty information, which is specially critical in medical diagnosis. This work proposes a 3D encoder-decoder architecture, based on V-Net \cite{vnet} which is trained with patching techniques to reduce memory consumption and decrease the effect of unbalanced data. We also introduce voxel-wise uncertainty, both epistemic and aleatoric using test-time dropout and data-augmentation respectively. Uncertainty maps can provide extra information to expert neurologists, useful for detecting when the model is not confident on the provided segmentation.