Uncertainty-Aware Multi-Source Retinal Fluid Segmentation in OCT
arXiv:2607.12212
The paper proposes an attention‑guided TransUNet that segments three retinal fluid types in OCT images from multiple scanners, using domain‑adaptive normalization and providing uncertainty estimates that highlight unreliable pixels for clinical triage.
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.
9 pages, 2 figures, 5 tables. Code, model weights, and REST inference API are available on GitHub and Zenodo