Kömen, Jonah https://orcid.org/0009-0009-4606-4695
de Jong, Edwin D. https://orcid.org/0000-0002-0679-5196
Hense, Julius https://orcid.org/0009-0007-1160-1636
Marienwald, Hannah
Dippel, Jonas https://orcid.org/0000-0002-0552-8977
Naumann, Philip https://orcid.org/0000-0002-6929-8568
Marcus, Eric
Ruff, Lukas https://orcid.org/0000-0002-9707-297X
Alber, Maximilian
Teuwen, Jonas https://orcid.org/0000-0002-1825-1428
Klauschen, Frederick https://orcid.org/0000-0002-9131-2389
Müller, Klaus-Robert https://orcid.org/0000-0002-3861-7685
Funding for this research was provided by:
Bundesministerium für Bildung und Forschung (01IS14013A-E, 01GQ1115, 01GQ0850, 01IS18056A, 01IS18025A, 13GW0744D, BIFOLD25B)
Korea University (RS-2019-II190079, RS-2024-00457882)
Article History
Received: 24 July 2025
Accepted: 18 May 2026
First Online: 11 June 2026
Competing interests
: E.D.J., J.D., L.R., and M.A. are employees of Aignostics. M.A., F.K., and K.-R.M. are co-founders of Aignostics. F.K. is lead medical advisor and board member of Aignostics. K.R.M. is lead technical advisor to Aignostics. E.D.J. was previously employed at kaiko.ai. Aignostics and kaiko.ai developed foundation models that were assessed in this study (Atlas, RudolfV, Kaiko ViT-B/8). J.T. is co-founder of Glass Medical AI. The remaining authors declare no competing interests.