MCU-Net: A framework towards uncertainty representations for decision support system patient referrals in healthcare contexts

TL;DR

This paper introduces MCU-Net, a deep learning framework with uncertainty estimation for medical image segmentation, enabling automated referrals for uncertain cases to improve trust and performance in healthcare decision support systems.

Contribution

It presents a novel framework combining U-Net with Monte Carlo Dropout for uncertainty estimation, incorporating a human-in-the-loop mechanism for patient-specific decision support.

Findings

MCU-Net with epistemic uncertainty improves automated segmentation accuracy.

Uncertainty thresholding effectively identifies cases needing human review.

Framework enhances trust and reliability in AI-assisted healthcare decisions.

Abstract

Incorporating a human-in-the-loop system when deploying automated decision support is critical in healthcare contexts to create trust, as well as provide reliable performance on a patient-to-patient basis. Deep learning methods while having high performance, do not allow for this patient-centered approach due to the lack of uncertainty representation. Thus, we present a framework of uncertainty representation evaluated for medical image segmentation, using MCU-Net which combines a U-Net with Monte Carlo Dropout, evaluated with four different uncertainty metrics. The framework augments this by adding a human-in-the-loop aspect based on an uncertainty threshold for automated referral of uncertain cases to a medical professional. We demonstrate that MCU-Net combined with epistemic uncertainty and an uncertainty threshold tuned for this application maximizes automated performance on an individual patient level, yet refers truly uncertain cases. This is a step towards uncertainty representations when deploying machine learning based decision support in healthcare settings.