USE-Evaluator: Performance Metrics for Medical Image Segmentation Models with Uncertain, Small or Empty Reference Annotations

TL;DR

This paper introduces USE-Evaluator, a new performance metric for medical image segmentation that accounts for uncertain, small, or empty reference annotations, addressing limitations of traditional metrics in clinical settings.

Contribution

The paper proposes a novel evaluation metric that better reflects clinical realities by handling uncertain, small, or empty annotations, and compares its behavior to standard metrics across datasets.

Findings

Traditional metrics are inadequate for uncertain or empty annotations.

USE-Evaluator provides more reliable performance assessment in challenging clinical cases.

The code for the new metric is publicly available for further research.

Abstract

Performance metrics for medical image segmentation models are used to measure the agreement between the reference annotation and the predicted segmentation. Usually, overlap metrics, such as the Dice, are used as a metric to evaluate the performance of these models in order for results to be comparable. However, there is a mismatch between the distributions of cases and difficulty level of segmentation tasks in public data sets compared to clinical practice. Common metrics fail to measure the impact of this mismatch, especially for clinical data sets that include low signal pathologies, a difficult segmentation task, and uncertain, small, or empty reference annotations. This limitation may result in ineffective research of machine learning practitioners in designing and optimizing models. Dimensions of evaluating clinical value include consideration of the uncertainty of reference annotations, independence from reference annotation volume size, and evaluation of classification of empty reference annotations. We study how uncertain, small, and empty reference annotations influence the value of metrics for medical image segmentation on an in-house data set regardless of the model. We examine metrics behavior on the predictions of a standard deep learning framework in order to identify metrics with clinical value. We compare to a public benchmark data set (BraTS 2019) with a high-signal pathology and certain, larger, and no empty reference annotations. We may show machine learning practitioners, how uncertain, small, or empty reference annotations require a rethinking of the evaluation and optimizing procedures. The evaluation code was released to encourage further analysis of this topic. https://github.com/SophieOstmeier/UncertainSmallEmpty.git