Research
Uncertainty-Aware Multi-Source Retinal Fluid Segmentation in OCT
arXiv:2607.12212v1 Announce Type: cross Abstract: Measuring retinal fluid from optical coherence tomography (OCT) drives treatment decisions in macular disease, but manual annotation is slow and segme
arXiv:2607.12212v1 Announce Type: cross Abstract: Measuring retinal fluid from optical coherence tomography (OCT) drives treatment decisions in macular disease, but manual annotation is slow and segmentation models trained on one scanner degrade on another. We present an attention-guided TransUNet that segments three fluid types across four independent OCT sources, combining a domain-adaptive normalisation scheme with an uncertainty estimate that flags unreliable pixels. The model reaches a mean fluid Dice of 0.78, and -- most usefully for clinicians -- its uncertainty is 1.34x higher exactly where expert graders disagree (p<10^-4), turning a raw segmentation map into an actionable clinical triage signal.
Related
- SAIL: Structure-Aware Interpretable Learning for Anatomy-Aligned Post-hoc Explanations in OCT
- Deep Learning-assisted AMD Staging based on OCT and OCT Angiography
- From Model Uncertainty to Human Attention: Localization-Aware Visual Cues for Scalable Annotation Review
- An Empirical Study on Variance-based MC Dropout Uncertainty-Error Correlation in 2D Brain Tumor Segmentation
Source: arXiv cs.CV | 2026-07-15