{"id":38997,"date":"2026-08-21T15:27:57","date_gmt":"2026-08-21T15:27:57","guid":{"rendered":"https:\/\/lamarr-institute.org\/publication\/knowing-when-not-to-predict-self-supervised-learning-and-abstention-for-safer-dr-screening\/"},"modified":"2026-09-17T11:17:53","modified_gmt":"2026-09-17T11:17:53","slug":"knowing-when-not-to-predict-self-supervised-learning-and-abstention-for-safer-dr-screening","status":"publish","type":"publication","link":"https:\/\/lamarr-institute.org\/de\/publication\/knowing-when-not-to-predict-self-supervised-learning-and-abstention-for-safer-dr-screening\/","title":{"rendered":"Knowing When Not to Predict: Self Supervised Learning and Abstention for Safer {DR} Screening"},"content":{"rendered":"<p>Self-supervised learning ({SSL}) is now a standard way to pretrain medical image models, but performance is still mostly judged by downstream accuracy. For safety-critical screening tasks such as diabetic retinopathy grading, this is not enough: a model must also know when its predictions are unreliable and defer uncertain cases for clinical review. In this work, we examine how the length of {SSL} pretraining influences calibrated confidence and confidence-based abstention. We evaluate multiple {SSL} checkpoints under a fixed fine-tuning protocol and assess calibrated confidence, coverage, selective accuracy, and selective macro-F1. Across datasets and data regimes, {SSL} pretraining improves selective prediction compared to training from scratch. Unlike prior {SSL} studies that primarily evaluate downstream accuracy or {AUROC}, we analyze how {SSL} pretraining duration influences confidence behavior under calibrated confidence-based abstention. However, once accuracy saturates, selective performance can still change markedly across checkpoints, and longer pretraining does not consistently improve reliability. These results underscore the importance of abstention-aware evaluation and suggest that pretraining length should be treated as an important reliability-related design choice rather than only a computational detail. Code is available at {GitHub}.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Self-supervised learning ({SSL}) is now a standard way to pretrain medical image models, but performance is still mostly judged by downstream accuracy. For safety-critical screening tasks such as diabetic retinopathy grading, this is not enough: a model must also know when its predictions are unreliable and defer uncertain cases for clinical review. In this work, we examine how the length of {SSL} pretraining influences calibrated confidence and confidence-based abstention. We [&hellip;]<\/p>\n","protected":false},"author":14,"featured_media":0,"template":"","meta":{"_acf_changed":false,"footnotes":""},"publication-type":[32],"class_list":["post-38997","publication","type-publication","status-publish","hentry","publication-type-inproceedings"],"acf":[],"publishpress_future_workflow_manual_trigger":{"enabledWorkflows":[]},"_links":{"self":[{"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication\/38997","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication"}],"about":[{"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/types\/publication"}],"author":[{"embeddable":true,"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/users\/14"}],"version-history":[{"count":0,"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication\/38997\/revisions"}],"wp:attachment":[{"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/media?parent=38997"}],"wp:term":[{"taxonomy":"publication-type","embeddable":true,"href":"https:\/\/lamarr-institute.org\/de\/wp-json\/wp\/v2\/publication-type?post=38997"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}