# U2-Net: A Bayesian U-Net model with epistemic uncertainty feedback for photoreceptor layer segmentation in pathological OCT scans

## TL;DR

This paper presents a Bayesian U-Net model for segmenting the photoreceptor layer in pathological OCT scans, providing accurate results and uncertainty maps to identify potential errors or pathologies.

## Contribution

The novel Bayesian U-Net architecture incorporates epistemic uncertainty feedback, enhancing segmentation accuracy and error detection in pathological OCT scans.

## Key findings

- Improved Dice index and precision/recall performance over baseline U-Net.
- Uncertainty maps effectively highlight areas of potential pathology or segmentation errors.
- Model performance inversely correlates with uncertainty estimates, aiding manual review.

## Abstract

In this paper, we introduce a Bayesian deep learning based model for segmenting the photoreceptor layer in pathological OCT scans. Our architecture provides accurate segmentations of the photoreceptor layer and produces pixel-wise epistemic uncertainty maps that highlight potential areas of pathologies or segmentation errors. We empirically evaluated this approach in two sets of pathological OCT scans of patients with age-related macular degeneration, retinal vein oclussion and diabetic macular edema, improving the performance of the baseline U-Net both in terms of the Dice index and the area under the precision/recall curve. We also observed that the uncertainty estimates were inversely correlated with the model performance, underlying its utility for highlighting areas where manual inspection/correction might be needed.