Uncertainty-Aware Last-Layer Adaptation of RETFound for Referable Diabetic Retinopathy Screening Under Dataset Shift

Researchers evaluated uncertainty-aware last-layer adaptation methods for referable diabetic retinopathy screening using RETFound, a self-supervised vision-transformer foundation model, on the APTOS 2019 and DDR datasets. The strongest method on APTOS deferred about 20% of cases and eliminated false negatives among accepted cases, but Bayesian methods did not uniquely reduce false negatives, and the SNGP checkpoint transferred poorly to DDR. The study underscores the need for safety-centered evaluation beyond aggregate accuracy for trustworthy retinal screening claims.

arXiv:2607.02569v1 Announce Type: new Abstract: This paper presents a safety-centered empirical evaluation of uncertainty-aware last-layer adaptation for referable diabetic retinopathy screening using RETFound, a self-supervised vision-transformer retinal foundation model used here as a frozen feature encoder, and the public APTOS 2019 and DDR diabetic retinopathy fundus image datasets. We compare a cached-feature softmax head, post-hoc temperature scaling, variational Bayesian last-layer heads, a diagonal Laplace last-layer approximation, and an SNGP-style cached-feature head. On APTOS, uncertainty-aware operating points improved sensitivity and selective-referral behavior. The strongest APTOS selective-referral result deferred approximately 20 percent of cases and reduced accepted-case false negatives to zero while preserving high accepted-case specificity. However, threshold tuning also reduced false negatives at high false-positive cost, so false-negative reduction alone was not unique to Bayesian modeling. On DDR, native Bayesian heads qualitatively reproduced the APTOS direction but with weaker tradeoffs, while the APTOS-trained SNGP checkpoint transferred poorly and failed to provide useful external selective-referral behavior. These results highlight the value of safety-centered evaluation beyond aggregate accuracy: uncertainty-aware last-layer heads can improve internal safety-oriented operating points, but trustworthy retinal screening claims require explicit safety-coverage evaluation and second-dataset validation under shift.