Bigger Isn't Always Better: Towards a General Prior for Medical Image Reconstruction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Glaszner, Lukas, Zach, Martin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916564761051136
author Glaszner, Lukas
Zach, Martin
author_facet Glaszner, Lukas
Zach, Martin
contents Diffusion model 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.
format Preprint
id arxiv_https___arxiv_org_abs_2501_07376
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Bigger Isn't Always Better: Towards a General Prior for Medical Image Reconstruction
Glaszner, Lukas
Zach, Martin
Image and Video Processing
Diffusion model 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.
title Bigger Isn't Always Better: Towards a General Prior for Medical Image Reconstruction
topic Image and Video Processing
url https://arxiv.org/abs/2501.07376