Multiparameter Uncertainty Mapping in Quantitative Molecular MRI using a Physics-Structured Variational Autoencoder (PS-VAE)
Alex Finkelstein, Ron Moneta, Or Zohar, Michal Rivlin, Moritz Zaiss, Dinora Friedmann Morvinski, and Or Perlman · 2026
Read Multiparameter Uncertainty Mapping in Quantitative Molecular MRI using a Physics-Structured Variational Autoencoder (PS-VAE) by Alex Finkelstein, Ron Moneta, Or Zohar, Michal Rivlin, Moritz Zaiss, Dinora Friedmann Morvinski, and Or Perlman (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-2602-03317.txt.
Description
Alex Finkelstein, leading 6 other authors, maps doubt in molecular MRI scans with a physics-structured variational autoencoder. MRI here means magnetic resonance imaging. Quantitative scans such as magnetic resonance fingerprinting aim to read tissue traits from signal changes over time. Most solvers for that inverse problem give point values with no clear doubt attached. That limits trust and the clinical uptake of quantitative biomarkers from scans.
Their autoencoder builds a differentiable spin physics simulator into self-supervised learning. It outputs a full posterior spread for each voxel, with covariance that captures how the traits correlate. Tests span lab phantoms, tumor-bearing mice, healthy volunteers, and a subject with glioblastoma. The posteriors matched brute-force Bayesian analysis while running orders of magnitude faster. Watching posteriors as signals accrue also informs protocol design.
Catalog record
| CREATOR | Alex Finkelstein, Ron Moneta, Or Zohar, Michal Rivlin, Moritz Zaiss, Dinora Friedmann Morvinski, and Or Perlman |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 46 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-03317