Denoising diffusion networks for normative modeling in neuroimaging Journal: Medical Image Analysis
Luke Whitbread, Lyle J. Palmer, Mark Jenkinson · 2026
Read Denoising diffusion networks for normative modeling in neuroimaging Journal: Medical Image Analysis by Luke Whitbread, Lyle J. Palmer, Mark Jenkinson (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-04886.txt.
Description
Brain imaging studies build reference ranges for each measure, so one person's score can be read against the norm. Luke Whitbread and 2 co-authors propose diffusion models as a single tool that estimates many measures at once. Standard pipelines fit one model per measure and drop the joint patterns between measures, which may carry real signal.
Two designs are tested: a modulated perceptron and a tabular transformer with two kinds of attention. On synthetic data and on UK Biobank brain measures, from two measures up to two hundred, the models stayed well calibrated per measure. The transformer held up far better as the count of measures grew and kept the joint structure.
Catalog record
| CREATOR | Luke Whitbread, Lyle J. Palmer, Mark Jenkinson |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 72 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-04886