Bayesian Generative Models for Knowledge Transfer in MRI Semantic Segmentation Problems

TL;DR

This paper introduces a Bayesian generative model for knowledge transfer in MRI segmentation, improving performance on small datasets by leveraging disease-related priors, outperforming traditional pre-training and random initialization methods.

Contribution

The paper presents a novel Generative Bayesian Prior network for effective knowledge transfer in MRI segmentation tasks with limited data.

Findings

Achieved higher Dice Similarity Coefficient on BRATS2018 small subsets.

Outperformed pre-train and random initialization baselines.

Demonstrated effectiveness of Bayesian priors in medical image segmentation.

Abstract

Automatic segmentation methods based on deep learning have recently demonstrated state-of-the-art performance, outperforming the ordinary methods. Nevertheless, these methods are inapplicable for small datasets, which are very common in medical problems. To this end, we propose a knowledge transfer method between diseases via the Generative Bayesian Prior network. Our approach is compared to a pre-train approach and random initialization and obtains the best results in terms of Dice Similarity Coefficient metric for the small subsets of the Brain Tumor Segmentation 2018 database (BRATS2018).