Uncertainty Quantification using Variational Inference for Biomedical Image Segmentation
This work addresses the need for interpretability and confidence in predictions for safety-critical biomedical applications, though it is incremental as it applies existing variational inference techniques to a specific domain.
The authors tackled the problem of uncertainty quantification in biomedical image segmentation by using a variational inference-based encoder-decoder architecture to segment brain tumors, achieving results evaluated on the BRATS dataset with Dice Similarity Coefficient and Intersection Over Union metrics.
Deep learning motivated by convolutional neural networks has been highly successful in a range of medical imaging problems like image classification, image segmentation, image synthesis etc. However for validation and interpretability, not only do we need the predictions made by the model but also how confident it is while making those predictions. This is important in safety critical applications for the people to accept it. In this work, we used an encoder decoder architecture based on variational inference techniques for segmenting brain tumour images. We evaluate our work on the publicly available BRATS dataset using Dice Similarity Coefficient (DSC) and Intersection Over Union (IOU) as the evaluation metrics. Our model is able to segment brain tumours while taking into account both aleatoric uncertainty and epistemic uncertainty in a principled bayesian manner.