Improving Uncertainty-based Out-of-Distribution Detection for Medical Image Segmentation

TL;DR

This paper evaluates uncertainty-based methods for detecting out-of-distribution inputs in medical image segmentation, highlighting the limitations of uncertainty alone and proposing joint anatomical and lesion segmentation as a more effective approach.

Contribution

It introduces a comprehensive evaluation of OOD detection methods in medical imaging and demonstrates that joint anatomical and lesion segmentation improves OOD detection over uncertainty-based methods.

Findings

Uncertainty-based methods often fail to detect OOD inputs.

Joint segmentation of anatomy and lesions enhances OOD detection.

Evaluation includes 14 diverse OOD sources.

Abstract

Deep Learning models are easily disturbed by variations in the input images that were not seen during training, resulting in unpredictable behaviours. Such Out-of-Distribution (OOD) images represent a significant challenge in the context of medical image analysis, where the range of possible abnormalities is extremely wide, including artifacts, unseen pathologies, or different imaging protocols. In this work, we evaluate various uncertainty frameworks to detect OOD inputs in the context of Multiple Sclerosis lesions segmentation. By implementing a comprehensive evaluation scheme including 14 sources of OOD of various nature and strength, we show that methods relying on the predictive uncertainty of binary segmentation models often fails in detecting outlying inputs. On the contrary, learning to segment anatomical labels alongside lesions highly improves the ability to detect OOD inputs.