EUVIMEDEUVIMED entdecken
VergleichsstudieOpen Access · CC BY 4.0

A Comparative Study of IVIM ‐ MRI Fitting Techniques in Glioma Grading: Conventional, Bayesian, and Voxel‐Wise and Spatially‐Aware Deep Learning Approaches

Misha P T Kaandorp iD, Andras Jakab iD, Christian Federau iD, Peter T While iD

Journal of Magnetic Resonance Imaging · 2026

DOI: 10.1002/jmri.70301 ↗

Kurz gefasst

Vergleichsstudie in der Zeitschrift Journal of Magnetic Resonance Imaging, 2026. Verfasst von Misha P T Kaandorp und 3 weiteren Autor:innen. Beteiligte Einrichtung: Center for MR Research, University Children's Hospital Zurich, Zurich, Switzerland. Hauptthemen laut MEDLINE: Glioma, Deep Learning, Diffusion Magnetic Resonance Imaging, Brain Neoplasms. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: EMDO Foundation, Vontobel Foundation, Anna Müller-Grocholski.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Intravoxel incoherent motion (IVIM) analysis of diffusion‐weighted MRI (DWI) provides microvascular perfusion and diffusion information. However, parameter estimation is limited by noise sensitivity, variability across fitting methods, and lack of standardization. Deep‐learning (DL)–based approaches, particularly spatially‐aware transformers, may improve robustness, but their clinical utility remains unexplored.

Purpose

To evaluate conventional, Bayesian, and DL–based IVIM fitting methods in glioma patients, focusing on tumor grading accuracy. Study Type Retrospective. Population Fractal‐noise‐based simulations and preoperative DWI from 20 glioma patients (5 Grade‐2, 3 Grade‐3, 12 Grade‐4). Sequence Spin‐echo echo‐planar DWI, 16 b values (0–900 s/mm 2 ), three orthogonal directions, 3 T. Assessment IVIM parameter maps were compared across least squares (LSQ), segmented (SEG), Bayesian shrinkage (BSP), and spatial‐homogeneity (FBM) priors, and DL–based methods, including IVIM‐NET, novel spatially‐aware transformers (NATTEN‐17), and a refined version (SA‐17). Simulation accuracy was evaluated using median absolute percentage error (MDAPE) and bias using median percentage error. In vivo data were visually assessed by the authors for noise suppression and structural preservation. Whole‐tumor diffusion coefficient ( D ), pseudo‐diffusion coefficient ( D *), and signal fraction ( f ) values were evaluated across tumor grades and for differentiating Grade‐4 from Grade‐2/3 tumors. Statistical Tests Mann–Whitney U tests for group comparisons; tumor grading performance using receiver operating characteristic–area under the curve (ROC‐AUC), and pairwise AUC differences using the DeLong test. Significance: p < 0.05.

Results

Transformer‐based methods achieved superior simulation accuracy, with significantly lower MDAPE for f and D * than all other approaches: NATTEN‐17 (5.91%, 13.31%), SA‐17 (7.73%, 13.66%), LSQ (21.95%, 54.34%), SEG (17.10%, 21.27%), BSP (12.35%, 22.79%), FBM (16.32%, 20.67%). In vivo, they provided superior visual quality and tumor delineation in f and D * maps, producing seemingly denoised versions of LSQ, while preserving tumor heterogeneity. The spatially‐aware transformers yielded consistently the highest ROC‐AUC values, particularly for D * (SA‐17: 0.78), significantly outperforming LSQ (0.62), SEG (0.58), FBM (0.62), and IVIM‐NET (0.71). Data Conclusion Transformer‐based model fitting has the potential to provide clinically valuable IVIM parameter estimates and improved tumor grading accuracy. Evidence Level 3. Technical Efficacy Stage 2.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

GliomaDeep LearningDiffusion Magnetic Resonance ImagingBrain NeoplasmsImage Interpretation, Computer-AssistedImage Processing, Computer-AssistedHumansBayes TheoremMaleFemaleNeoplasm GradingMiddle AgedRetrospective StudiesAdultAgedMotionAlgorithmsComputer SimulationBrainReproducibility of Results

Stichwörter der Autor:innen

Bayesian model fittingdeep learning parameter estimationglioma tumor gradingintravoxel incoherent motion (IVIM)spatially‐aware transformer networks

Autor:innen und Institutionen

  • Center for MR Research, University Children's Hospital Zurich, Zurich, Switzerland.
  • University of Zurich, Zurich, Switzerland.
  • AI Medical AG, Zollikon, Switzerland.
  • Kyoto Prefectural University of Medicine, Kyoto, Japan.
  • Department of Radiology and Nuclear Medicine, St. Olav's University Hospital, Trondheim, Norway.
  • Department of Circulation and Medical Imaging, NTNU-Norwegian University of Science and Technology, Trondheim, Norway.

ORCID-Profile: Misha P T Kaandorp, Andras Jakab, Christian Federau, Peter T While

Förderung

  • EMDO Foundation
  • Vontobel Foundation
  • Anna Müller-Grocholski
  • Prof. Max Cloetta Foundations
  • Research Council of Norway
  • Swiss National Science Foundation
  • Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung
  • EMDO Stiftung

Zitieren

APA

Kaandorp, M. P. T., Jakab, A., Federau, C., & While, P. T. (2026). A Comparative Study of IVIM ‐ MRI Fitting Techniques in Glioma Grading: Conventional, Bayesian, and Voxel‐Wise and Spatially‐Aware Deep Learning Approaches. Journal of Magnetic Resonance Imaging. https://doi.org/10.1002/jmri.70301

Vancouver

Kaandorp MPT, Jakab A, Federau C, While PT. A Comparative Study of IVIM ‐ MRI Fitting Techniques in Glioma Grading: Conventional, Bayesian, and Voxel‐Wise and Spatially‐Aware Deep Learning Approaches. Journal of Magnetic Resonance Imaging. 2026. doi:10.1002/jmri.70301

BibTeX
@article{Kaandorp2026,
  author = {Kaandorp, Misha P T and Jakab, Andras and Federau, Christian and While, Peter T},
  title = {A Comparative Study of IVIM ‐ MRI Fitting Techniques in Glioma Grading: Conventional, Bayesian, and Voxel‐Wise and Spatially‐Aware Deep Learning Approaches},
  journal = {Journal of Magnetic Resonance Imaging},
  year = {2026},
  doi = {10.1002/jmri.70301}
}
Kostenlos starten →