EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Self‐Supervised Deep Learning Framework for Rician Distribution Based Denoising and Modeling of Multi‐b Prostate Diffusion MRI

Mustafa Abbas iD, Wenyin Zhou, Stephan E Maier, Fredrik Langkilde, Stefan Kuczera iD

Magnetic Resonance in Medicine · 2026-08 · 96 (2) · 928-942

DOI: 10.1002/mrm.70373 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Magnetic Resonance in Medicine, online erschienen am 19.04.2026. Verfasst von Mustafa Abbas und 4 weiteren Autor:innen. Beteiligte Einrichtung: Department of Radiology, Institute of Clinical Sciences, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden. Hauptthemen laut MEDLINE: Diffusion Magnetic Resonance Imaging, Prostate, Deep Learning, Prostatic Neoplasms. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: The Swedish Cancer Society, The Swedish Prostate Cancer Federation, Assar Gabrielsson Foundation.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Purpose

Convolutional neural networks (CNNs) are evaluated for improved and accelerated denoising and Rician bias correction in multi‐b DW images with simultaneous signal modeling.

Methods

Prostate diffusion images from 46 individuals acquired at 20 linearly distributed b‐values ( without averaging were used for self‐supervised CNN training. CNNs were trained to output model parameter maps, from which DW images and ADC maps were synthesized, and to account for Rician bias. CNN architectures included were: conventional U‐Net, Attention U‐Net, and Residual Attention U‐Net with biexponential, kurtosis, and gamma distribution as signal models. Moreover, approaches without Rician bias correction and with noise maps as additional input were explored.

Results

For all signal models, synthetic DW images generated from CNN output are of similar quality compared to OBSIDIAN, a model‐based method employing iterative pixel‐wise fitting. Computed ADC and parameter maps obtained with CNN models were less noisy than those obtained with OBSIDIAN. Good quantitative agreement with OBSIDIAN in ROI‐averaged ADC values and, to a slightly lesser extent, kurtosis model parameters was observed. Results with omitted Rician bias correction clearly deviate, while the impact of the noise map input for the CNNs is less pronounced. Computation time was reduced to seconds compared to several hours for OBSIDIAN.

Conclusion

The CNN‐based method has potential in clinical use for achieving higher quality DW images and biomarker maps with drastically reduced computation time. Further refinement and larger training datasets are needed to achieve better generalizability and increased robustness.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Diffusion Magnetic Resonance ImagingProstateDeep LearningProstatic NeoplasmsImage Processing, Computer-AssistedSupervised Machine LearningHumansMaleConvolutional Neural NetworksSignal-To-Noise RatioAlgorithmsImage Interpretation, Computer-Assisted

Stichwörter der Autor:innen

ADCRician bias correctiondenoisingdiffusionprostate

Autor:innen und Institutionen

  • Department of Radiology, Institute of Clinical Sciences, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden.
  • Department of Intelligent Systems, School of Electrical Engineering and Computer Science, KTH Royal Institute of Technology, Stockholm, Sweden.
  • Department of Radiology, Brigham Women's Hospital, Harvard Medical School, Boston, MA, USA.

ORCID-Profile: Mustafa Abbas, Stefan Kuczera

Förderung

  • The Swedish Cancer Society
  • The Swedish Prostate Cancer Federation
  • Assar Gabrielsson Foundation
  • Stiftelsen Jubileumsklinikens Forskningsfond mot Cancer
  • The Swedish Research Council
  • Swedish governmental funding of clinical research (ALF)

Zitieren

APA

Abbas, M., Zhou, W., Maier, S. E., Langkilde, F., & Kuczera, S. (2026). Self‐Supervised Deep Learning Framework for Rician Distribution Based Denoising and Modeling of Multi‐b Prostate Diffusion MRI. Magnetic Resonance in Medicine, 96(2), 928-942. https://doi.org/10.1002/mrm.70373

Vancouver

Abbas M, Zhou W, Maier SE, Langkilde F, Kuczera S. Self‐Supervised Deep Learning Framework for Rician Distribution Based Denoising and Modeling of Multi‐b Prostate Diffusion MRI. Magnetic Resonance in Medicine. 2026;96(2):928-942. doi:10.1002/mrm.70373

BibTeX
@article{Abbas2026,
  author = {Abbas, Mustafa and Zhou, Wenyin and Maier, Stephan E and Langkilde, Fredrik and Kuczera, Stefan},
  title = {Self‐Supervised Deep Learning Framework for Rician Distribution Based Denoising and Modeling of Multi‐b Prostate Diffusion MRI},
  journal = {Magnetic Resonance in Medicine},
  year = {2026},
  volume = {96},
  number = {2},
  pages = {928--942},
  doi = {10.1002/mrm.70373}
}
Kostenlos starten →