Harnessing Uncertainty in Domain Adaptation for MRI Prostate Lesion Segmentation

TL;DR

This paper introduces a method for domain adaptation in MRI prostate lesion segmentation that models uncertainty to generate multiple target domain outputs from a single source, improving segmentation performance.

Contribution

It addresses the challenge of adapting from mp-MRI to VERDICT MRI by explicitly modeling uncertainty, enabling multiple outputs and better representations for improved segmentation.

Findings

Multiple outputs improve segmentation accuracy

Uncertainty modeling enhances domain adaptation

Method outperforms deterministic baselines

Abstract

The need for training data can impede the adoption of novel imaging modalities for learning-based medical image analysis. Domain adaptation methods partially mitigate this problem by translating training data from a related source domain to a novel target domain, but typically assume that a one-to-one translation is possible. Our work addresses the challenge of adapting to a more informative target domain where multiple target samples can emerge from a single source sample. In particular we consider translating from mp-MRI to VERDICT, a richer MRI modality involving an optimized acquisition protocol for cancer characterization. We explicitly account for the inherent uncertainty of this mapping and exploit it to generate multiple outputs conditioned on a single input. Our results show that this allows us to extract systematically better image representations for the target domain, when used in tandem with both simple, CycleGAN-based baselines, as well as more powerful approaches that integrate discriminative segmentation losses and/or residual adapters. When compared to its deterministic counterparts, our approach yields substantial improvements across a broad range of dataset sizes, increasingly strong baselines, and evaluation measures.