Enhancing computation speed and accuracy in deep image prior‐based parameter mapping
Magnetic Resonance in Medicine · 2025
DOI: 10.1002/mrm.30630 ↗Kurz gefasst
Fachartikel in der Zeitschrift Magnetic Resonance in Medicine, 2025. Verfasst von Max Hellström und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Diagnostics and Intervention, Umeå University, Umeå, Sweden. Hauptthemen laut MEDLINE: Magnetic Resonance Imaging, Image Processing, Computer-Assisted, Deep Learning. Die Arbeit nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Cancerforskningsfonden i Norrland, Lions Cancerforskningsfond, Region Västerbotten (Bas ALF).
Abstract (Originalsprache)
Purpose
To make Deep Image Prior (DIP)‐based parameter mapping faster, more accurate, and suitable for clinical applications, with added support for multislice and 3D datasets.
Methods
DIP leverages the inherent structure of an untrained image generator to address various inverse imaging tasks, including denoising. In this study, we enhance DIP‐based denoising for parameter mapping with warm‐start across neighboring image slices and different patient subjects. This approach leverages spatial similarity to reduce computation time. Additionally, we introduce an early‐stopping criterion that selects the denoising level based on MRI signal noise. We further investigate uncertainty calibration through dropout probability tuning to address issues with miscalibrated uncertainty estimates from Monte Carlo dropout. Furthermore, we explore reducing computation time by tuning learning rates and network complexity.
Results
We show that reusing image generator weights with warm‐start significantly accelerates the denoising of large datasets, reducing computation time by 78% to 95% across various tasks. The early stopping approach proved effective, eliminating the need to manually select the number of optimization steps. Dropout probability tuning helps mitigate the issue of miscalibrated uncertainty, though further refinements are necessary, particularly to achieve better calibration on a per‐pixel level. Additionally, tuning learning rates and network complexity provided valuable insights into optimizing the model for different tasks.
Conclusion
The proposed developments enable DIP‐based parameter mapping to become faster, more accurate, and, consequently, more practical and scalable for clinical applications involving larger datasets.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Diagnostics and Intervention, Umeå University, Umeå, Sweden.
- Department of Computing Science, Umeå University, Umeå, Sweden.
ORCID-Profile: Max Hellström, Polina Kurtser, Tommy Löfstedt, Anders Garpebring
Förderung
- Cancerforskningsfonden i Norrland
- Lions Cancerforskningsfond
- Region Västerbotten (Bas ALF)
- The Swedish Research Council
Zitieren
APA
Hellström, M., Kurtser, P., Löfstedt, T., & Garpebring, A. (2025). Enhancing computation speed and accuracy in deep image prior‐based parameter mapping. Magnetic Resonance in Medicine. https://doi.org/10.1002/mrm.30630
Vancouver
Hellström M, Kurtser P, Löfstedt T, Garpebring A. Enhancing computation speed and accuracy in deep image prior‐based parameter mapping. Magnetic Resonance in Medicine. 2025. doi:10.1002/mrm.30630
BibTeX
@article{Hellstrm2025,
author = {Hellström, Max and Kurtser, Polina and Löfstedt, Tommy and Garpebring, Anders},
title = {Enhancing computation speed and accuracy in deep image prior‐based parameter mapping},
journal = {Magnetic Resonance in Medicine},
year = {2025},
doi = {10.1002/mrm.30630}
}
EUVIMED