Enhancing Uncertainty Estimation in Semantic Segmentation via Monte-Carlo Frequency Dropout

TL;DR

This paper introduces a frequency domain extension to MC Dropout for semantic segmentation, improving uncertainty estimation and boundary accuracy in medical imaging tasks.

Contribution

It proposes a novel MC-Frequency Dropout method that attenuates signal frequencies during inference to better estimate uncertainties in segmentation.

Findings

Improves calibration and convergence in segmentation models.

Enhances boundary delineation and uncertainty estimation.

Shows effectiveness across multiple medical imaging modalities.

Abstract

Monte-Carlo (MC) Dropout provides a practical solution for estimating predictive distributions in deterministic neural networks. Traditional dropout, applied within the signal space, may fail to account for frequency-related noise common in medical imaging, leading to biased predictive estimates. A novel approach extends Dropout to the frequency domain, allowing stochastic attenuation of signal frequencies during inference. This creates diverse global textural variations in feature maps while preserving structural integrity -- a factor we hypothesize and empirically show is contributing to accurately estimating uncertainties in semantic segmentation. We evaluated traditional MC-Dropout and the MC-frequency Dropout in three segmentation tasks involving different imaging modalities: (i) prostate zones in biparametric MRI, (ii) liver tumors in contrast-enhanced CT, and (iii) lungs in chest X-ray scans. Our results show that MC-Frequency Dropout improves calibration, convergence, and semantic uncertainty, thereby improving prediction scrutiny, boundary delineation, and has the potential to enhance medical decision-making.