Abstract
Medical foundation models learn latent representations of clinically meaningful phenotypes, yet their ability to support controllable image generation remains largely unexplored. We evaluate four retinalfoundationmodelswithintherepresentationtokenizerframework and examine whether demographic and clinical information encoded in latent representations from foundation models is preserved during synthetic image generation. We show that generated representations and images faithfully inherit phenotype information when evaluated within theiroriginatingfoundationmodels,consistentlyoutperformingconventionallatentdiffusiononmultipledownstreampredictiontasks.However, thesegainslargelydisappearwhenevaluatedusingclassifierstrainedon realimages,revealingapreviouslyuncharacterisedsynthetic-to-realrepresentation gap. These findings demonstrate that foundation-model latentspacesprovideapowerfulsubstrateforcontrollableretinalsynthesis whilehighlightingtheneedtobetteralignsyntheticrepresentationswith real-image distributions.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/SASHIMI_018.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: Not Submitted
Link to Open Review
Open Review Page: https://openreview.net/forum?id=0Wt6mwwRlH
BibTex
@InProceedings{WakZuz_Evaluation_MICCAISAT2026,
author = { Wakefield-Skórniewska, Zuzanna A. AND Papież, Bartłomiej W.},
title = { { Evaluation of Clinically Steerable Retinal Image Generation from Foundation Model Latent Spaces } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS 17258},
month = {pending},
page = {pending}
}
