@inproceedings{39f071077b07462383d273e283904180,
title = "Bigger Isn{\textquoteright}t Always Better: Towards a General Prior for Medical Image Reconstruction",
abstract = "Diffusion models have been successfully applied to many inverse problems, including MRI and CT reconstruction. Researchers typically re-purpose models originally designed for unconditional sampling without modifications. Using two different posterior sampling algorithms, we show empirically that such large networks are not necessary. Our smallest model, effectively a ResNet, performs almost as good as an attention U-Net on in-distribution reconstruction, while being significantly more robust towards distribution shifts. Furthermore, we introduce models trained on natural images and demonstrate that they can be used in both MRI and CT reconstruction, out-performing model trained on medical images in out-of-distribution cases. As a result of our findings, we strongly caution against simply re-using very large networks and encourage researchers to adapt the model complexity to the respective task. Moreover, we argue that a key step towards a general diffusion-based prior is training on natural images.",
keywords = "Data-driven prior, Medical image reconstruction, Score-based model",
author = "Lukas Glaszner and Martin Zach",
note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.; 46th Annual Conference of the German Association for Pattern Recognition, DAGM-GCPR 2024 ; Conference date: 10-09-2024 Through 13-09-2024",
year = "2025",
month = apr,
day = "23",
doi = "10.1007/978-3-031-85181-0\_18",
language = "English",
isbn = "9783031851803",
series = "Lecture Notes in Computer Science",
publisher = "Springer, Cham",
pages = "275--291",
editor = "Daniel Cremers and Zorah L{\"a}hner and Michael Moeller and Matthias Nie{\ss}ner and Bj{\"o}rn Ommer and Rudolph Triebel",
booktitle = "Pattern Recognition - 46th DAGM German Conference, DAGM GCPR 2024, Proceedings",
}