MRI brain tumor segmentation and uncertainty estimation using 3D-UNet architectures

TL;DR

This paper presents a 3D-UNet based approach for brain tumor segmentation in MRI that incorporates uncertainty estimation and ensemble techniques to improve accuracy and reliability in medical diagnosis.

Contribution

It introduces a hybrid 3D-UNet architecture with uncertainty estimation using TTD and TTA, and demonstrates improved performance on the BraTS'20 Challenge.

Findings

Ensemble models outperform individual models in segmentation accuracy.

Uncertainty estimation provides valuable information for medical diagnosis.

Hybrid approach increases segmentation precision.

Abstract

Automation of brain tumor segmentation 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 especially critical in medical diagnosis. This work studies 3D encoder-decoder architectures trained with patch-based techniques to reduce memory consumption and decrease the effect of unbalanced data. The different trained models are then used to create an ensemble that leverages the properties of each model, thus increasing the performance. We also introduce voxel-wise uncertainty information, both epistemic and aleatoric using test-time dropout (TTD) and data-augmentation (TTA) respectively. In addition, a hybrid approach is proposed that helps increase the accuracy of the segmentation. The model and uncertainty estimation measurements proposed in this work have been used in the BraTS'20 Challenge for task 1 and 3 regarding tumor segmentation and uncertainty estimation.