Uncertainty-Aware Training for Cardiac Resynchronisation Therapy Response Prediction

TL;DR

This paper introduces an uncertainty-aware training approach for deep learning models predicting cardiac resynchronisation therapy response, aiming to improve interpretability and confidence in clinical predictions by quantifying and incorporating uncertainty during training.

Contribution

It quantifies data and model uncertainty in cardiac therapy prediction and proposes a novel uncertainty-aware loss function to enhance model confidence and trustworthiness.

Findings

Increased confidence in true positive predictions.

Evidence of reduced false negative confidence.

Promising initial results in uncertainty quantification.

Abstract

Evaluation of predictive deep learning (DL) models beyond conventional performance metrics has become increasingly important for applications in sensitive environments like healthcare. Such models might have the capability to encode and analyse large sets of data but they often lack comprehensive interpretability methods, preventing clinical trust in predictive outcomes. Quantifying uncertainty of a prediction is one way to provide such interpretability and promote trust. However, relatively little attention has been paid to how to include such requirements into the training of the model. In this paper we: (i) quantify the data (aleatoric) and model (epistemic) uncertainty of a DL model for Cardiac Resynchronisation Therapy response prediction from cardiac magnetic resonance images, and (ii) propose and perform a preliminary investigation of an uncertainty-aware loss function that can be used to retrain an existing DL image-based classification model to encourage confidence in correct predictions and reduce confidence in incorrect predictions. Our initial results are promising, showing a significant increase in the (epistemic) confidence of true positive predictions, with some evidence of a reduction in false negative confidence.