UATTA-ENS: Uncertainty Aware Test Time Augmented Ensemble for PIRC Diabetic Retinopathy Detection

TL;DR

This paper introduces UATTA-ENS, a novel ensemble method that incorporates test-time augmentation to produce well-calibrated and uncertainty-aware predictions for diabetic retinopathy classification, enhancing diagnostic reliability.

Contribution

The paper proposes UATTA-ENS, a new ensemble approach that integrates test-time augmentation to improve uncertainty estimation and calibration in diabetic retinopathy detection.

Findings

UATTA-ENS provides better calibration than traditional ensembles.

The method improves the reliability of predictions in medical diagnosis.

It achieves comparable or superior accuracy to existing methods.

Abstract

Deep Ensemble Convolutional Neural Networks has become a methodology of choice for analyzing medical images with a diagnostic performance comparable to a physician, including the diagnosis of Diabetic Retinopathy. However, commonly used techniques are deterministic and are therefore unable to provide any estimate of predictive uncertainty. Quantifying model uncertainty is crucial for reducing the risk of misdiagnosis. A reliable architecture should be well-calibrated to avoid over-confident predictions. To address this, we propose a UATTA-ENS: Uncertainty-Aware Test-Time Augmented Ensemble Technique for 5 Class PIRC Diabetic Retinopathy Classification to produce reliable and well-calibrated predictions.