Aleatoric uncertainty in accelerated prostate MRI reconstruction: echo-train dropout versus Gaussian noise Monte Carlo sampling
Magnetic Resonance Materials in Physics, Biology and Medicine · 26.05.2026
DOI: 10.1007/s10334-026-01369-0 ↗Kurz gefasst
Fachartikel in der Zeitschrift Magnetic Resonance Materials in Physics, Biology and Medicine, online erschienen am 26.05.2026. Verfasst von Quintin van Lohuizen und 6 weiteren Autor:innen. Beteiligte Einrichtung: University Medical Center Groningen, Hanzeplein 1, 9713 GZ, Groningen, the Netherlands. q.y.van.lohuizen@umcg.nl. Die Arbeit nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Siemens Healthineers, Health~Holland.
Abstract (Originalsprache)
Purpose
Although undersampling combined with deep learning (DL)-based reconstruction shortens MRI acquisition, it increases the chance of inaccuracies, highlighting the need for quantifiable uncertainty measures. Two inference-time perturbation strategies, echo-train dropout (ET-Drop) and Gaussian noise Monte Carlo sampling (GN-MC), were compared in terms of the correlation between their variance-based uncertainty maps and absolute reconstruction error in DL-accelerated T2w prostate MRI.
Methods
This retrospective multi-center study used a publicly available dataset with 312 k-spaces from NYU for training and a dataset with 120 k-spaces from University Medical Center Groningen for external validation. Fully sampled 3 T data were retrospectively undersampled to acceleration factors R = 3 and R = 6 and reconstructed by a vSHARP model. Per slice, five GN-MC perturbations were reconstructed by adding complex noise at 2.5σ, and five ET-Drop perturbations, created by omitting non-central echo trains. Voxel-wise aleatoric uncertainty was defined as the variance (σ 2 ) across these reconstructions and correlated with absolute reconstruction error over whole slices and within the prostate.
Results
Both uncertainties yielded moderate slice-level correlations with absolute error. At R = 3, ET-Drop slightly outperformed GN-MC (median ρ = 0.39 vs 0.35; p < 0.001). At R = 6, the ranking reversed (0.44 vs 0.40; p < 0.001). Correlations within the prostate fell to 0.10–0.15. ET-Drop variance maps were dominated by coil sensitivities.
Conclusion
Both perturbation strategies yield variance-based uncertainty maps that correlate moderately with voxel-wise error. More importantly, they consistently highlighted acquisition-related fragility, supporting the role of uncertainty mapping as a useful quality-control tool in prostate MRI.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- University Medical Center Groningen, Hanzeplein 1, 9713 GZ, Groningen, the Netherlands. q.y.van.lohuizen@umcg.nl.
- , Piet Fransenlaan 127, 9713WN, Groningen, the Netherlands. q.y.van.lohuizen@umcg.nl.
- University Medical Center Groningen, Hanzeplein 1, 9713 GZ, Groningen, the Netherlands.
- Radboud University Medical Center, Geert Grooteplein Zuid 10, 6525 GA, Nijmegen, the Netherlands.
- University of Twente, Drienerlolaan 5, 7522 NB, Enschede, the Netherlands.
- Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX, Amsterdam, the Netherlands.
Förderung
- Siemens Healthineers
- Health~Holland
Zitieren
APA
van Lohuizen, Q., Fransen, S. J., Huisman, H., Wolterink, J. M., Kwee, T. C., Yakar, D., & Simonis, F. F. J. (2026). Aleatoric uncertainty in accelerated prostate MRI reconstruction: echo-train dropout versus Gaussian noise Monte Carlo sampling. Magnetic Resonance Materials in Physics, Biology and Medicine. https://doi.org/10.1007/s10334-026-01369-0
Vancouver
van Lohuizen Q, Fransen SJ, Huisman H, Wolterink JM, Kwee TC, Yakar D, et al. Aleatoric uncertainty in accelerated prostate MRI reconstruction: echo-train dropout versus Gaussian noise Monte Carlo sampling. Magnetic Resonance Materials in Physics, Biology and Medicine. 2026. doi:10.1007/s10334-026-01369-0
BibTeX
@article{vanLohuizen2026,
author = {van Lohuizen, Quintin and Fransen, Stefan J and Huisman, Henkjan and Wolterink, Jelmer M and Kwee, Thomas C and Yakar, Derya and Simonis, Frank F J},
title = {Aleatoric uncertainty in accelerated prostate MRI reconstruction: echo-train dropout versus Gaussian noise Monte Carlo sampling},
journal = {Magnetic Resonance Materials in Physics, Biology and Medicine},
year = {2026},
doi = {10.1007/s10334-026-01369-0}
}
EUVIMED