Cape Town - 2026 ISMRM-ISMRT Annual Meeting and Exhibition - Quantitative Mapping: Acquisition and Reconstruction

Quantitative Mapping: Acquisition and Reconstruction

Digital Poster

Acquisition & Reconstruction

Monday, 11 May 2026

Digital Posters Row A

16:10 - 17:05

Session Number: 360-04

No CME/CE Credit

Similar Sessions

Acquisition and Reconstruction Methods for Quantitative MRI

		360-04-001. Phase-cycled bSSFP-based relaxometry and susceptibility mapping with periodicity-informed parameter estimation (PIPE) Berk Can Acikgoz, Eva Peper, Gabriele Bonanno, Tom Hilbert, Suzanne Anderson, Jessica Bastiaansen Inselspital, Bern University Hospital, University of Bern, Switzerland Impact: The PIPE method improved the robustness (~40% RMSE reduction) of phase-cycled bSSFP–based quantitative mapping to SNR loss from acceleration, enabling higher acceleration without compromising accuracy and precision, supporting clinical translation of bSSFP’s multiparametric capabilities for practical quantitative imaging applications.
		360-04-002. 3D Mapping of Transverse Relaxation Parameters in the Human Brain With Self-Corrected AUSFIDE Eunseo Bae, Sungsuk Oh, Wonjun Son, Nasrat Niva, Felix Wehrli, Hyunyeol Lee Kyungpook National University, Daegu, Korea, Republic of Impact: Self-corrected AUSFIDE, a calibration-free technique for mitigating errors from field drifts and k-space trajectory mismatch in AUSFIDE, was evaluated via in vivo brain scans. The new method achieves rapid and robust 3D quantification of the human brain transverse relaxation parameters.
		360-04-003. B1+ Insensitive Quantitative Multiparametric Mapping in the Human Brain at 7T Using Phase-Cycled bSSFP Celik Boga, Berk Acikgoz, Nils Plähn, Jessica Bastiaansen, Anke Henning University of Texas Southwestern Medical Center, Dallas, United States of America Impact: With this novel approach,$B_1^+$-insensitive, quantitative, whole brain $T_1, T_2,$proton density and $\Delta{}B_0$ maps are obtained with 1 mm³ isotropic resolution at 7T within 8 minutes, easing the utilization of multiparametric imaging in clinical applications .
		360-04-004. Physics-guided Neural Network for Quantitative Parameter Mapping using Balanced Steady State Free Precession MRI HyeRyeong Choi, Huan Luu, Sung-Hong Park Korea Advanced Institute of Science & Technology, Daejeon, Korea, Republic of Impact: This study improves quantitative bSSFP mapping accuracy and efficiency with only simulated data, reducing cost. It provides a framework for reliable parameter estimation and data synthesis, enabling integration of advanced modeling methods and facilitating data augmentation for developing MRI techniques.
		360-04-005. Physics-reinforced Implicit Neural Representation for Scan-specific Multiparametric qMRI Reconstruction Ruimin Feng, Albert Jang, Xingxin He, Fang Liu Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital and Harvard Medical School, Charlestown, United States of America Impact: The proposed method enables accurate and efficient scan-specific multiparametric qMRI reconstruction by integrating implicit neural representation, physics-based modeling, and model adaptation. The framework requires no fully sampled data, improving the practicality of quantitative imaging for both research and clinical applications.
		360-04-006. Complex-Valued MR Fingerprinting (CV-MRF) for Simultaneous Estimation of R2, Field Perturbations, and Transceive Phase Matthew Cherukara, Kevin McNally, Karin Shmueli, Patrick Fuchs University College London, London, United Kingdom Impact:* We propose a novel dictionary matching method for field, transceive phase and $R_2^*$ mapping, validated using simulated multi-echo gradient echo data and demonstrated in-vivo. This approach may help improve precision in this pre-processing step fundamental for QSM, EPT, and beyond.
		360-04-007. Accelerated High-Resolution T1-Mapping of Human Embryos: Validating Multicontrast-Recon Resolution and Accuracy Kazuma Iwazaki, Naoto Fujita, Shigehito YAMADA, Yasuhiko Terada The University of Tsukuba, Tsukuba, Japan Impact: This work makes high-resolution T1 mapping of human embryos practical (e.g., AF=8). Researchers can now move beyond morphology to quantitatively analyze biophysical tissue development, enabling the creation of new high-resolution quantitative atlases of human embryogenesis.
		360-04-008. SSL-MIMOSA: Self-Supervised Learning for Fast Multiparameter Estimation Including T₂* Mapping in Quantitative MRI with MIMOSA Taewoong Lee, Yohan Jun, Yuting Chen, Borjan Gagoski, Gabriel Ramos Llordén, Berkin Bilgic Harvard University, Cambridge, United States of America Impact: SSL-MIMOSA estimates five quantitative maps, including T₂*, from a single scan. It replaces large dictionary matching with a self-supervised learning framework, enables reconstruction within minutes through transfer learning, and generalizes to 7T, advancing practical, comprehensive qMRI toward routine clinical/research use.
		360-04-009. Open-Source Quantitative MRI: Full Implementation of Acquisition and Reconstruction in BART Daniel Mackner, Philip Schaten, Markus Huemer, Vitali Telezki, Moritz Blumenthal, Xiaoqing Wang, Martin Uecker Graz University of Technology, Graz, Austria Impact: By providing open-source sequence definitions and reconstruction, advanced quantitative MRI methods can be fully reproduced. This opens the door for collaborative improvements and comparison of quantitative MRI techniques.
		360-04-010. Different Optimal AI Acceleration Settings on Different Scanners for Quantitative Measurements Using 3D-QALAS Maarten Naeyaert, PETER VAN SCHUERBEEK, Manon Roose, Hubert Raeymaekers Universitair Ziekenhuis Brussel, Brussels, Belgium Impact: AI reconstruction algorithms focus on image quality, not the reproducibility of quantitative measurements. Strategies for accelerating 3D-QALAS acquisitions were investigated on two vendors. Different optimal acceleration settings should be used to achieve the best quality, depending on the vendor.
		360-04-011. Synthetic Image and Multi-coil K-space Data Generation From Multi-parameter Maps Boran Kilic, Nikolaus Weiskopf, Kerrin Pine Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany Impact: Physics-consistent synthesis of images and multi-coil k-space from MPMs in high-resolution ultra-high-field qMRI enables scalable dataset creation for machine learning applications, reducing the need for acquiring real data and providing ground-truth tailored to sequence parameters, sensitivity maps, and noise models.
		360-04-012. Extending an MRI simulator for education (eduMRIsim) with realistically simulated artefacts Stephanie Gonzalez Riedel, Marcel Breeuwer, Alexander Raaijmakers Eindhoven University of Technology, Eindhoven, Netherlands Impact: eduMRIsim enables students to interactively explore how scan parameters affect image contrast, artefacts and SNR. This offers students the opportunity to put theory into practice and gain a deeper understanding of MR physics without requiring clinical scanner access.
		360-04-013. Variational Reconstruction Networks Preserving Phase-cycled bSSFP Signal Properties for T1 and T2 Mapping in the Brain David Sapienza, Nicholas Simic, Sam Ruysseveldt, Berk Acikgoz, Li Feng, Gabriele Bonanno, Kelvin Chow, Jessica Bastiaansen, Eva Peper Swiss Institute for Translational and Entrepreneurial Medicine, Bern, Switzerland, Switzerland Impact: Phase-cycled bSSFP offers high SNR and simultaneous T1/T2 mapping. Reconstruction of accelerated data requires preserving the signal profile for quantitative mapping. The proposed variational networks enable fast reconstruction and accurate mapping by incorporating the phase-cycling dimension in their internal representation.
		360-04-014. Fast, Differentiable Physics Models via Interpolation on Smoothed Manifolds: An Application to MRF Imraj Singh, Andrew Dupuis, Mark Griswold Case Western Reserve University, Cleveland, United States of America Impact: This study validates reformulating quantised physics models into fast, differentiable surrogates. This crucial step enables applying powerful gradient-based methods, like uncertainty quantification and physics-informed deep learning, to a broad new class of complex physics problems.
		360-04-015. Phase-only, but not Complex or Magnitude Fitting, Provides Optimal Robustness for RF Phase-Modulated GRE T2 Mapping Daiki Tamada, Diego Hernando, Scott Reeder University of Wisconsin - Madison, Madison, United States of America Impact: Theoretical and experimental analyses reveal that RF phase-modulated GRE T2 mapping using the phase-only estimation provides rapid and accurate T2 measurements with robustness to T1 and B1+ variations.
		360-04-016. Uncertainty Quantification in Dictionary Matching: an Efficient Framework for Statistically Interpretable Quantitative MRI Brian Toner, Ute Goerke, Eze Ahanonu, Kevin Johnson, Vibhas Deshpande, Holden Wu, Maria Altbach, Ali Bilgin University of Arizona, Tucson, United States of America Impact: Dictionary matching is ubiquitous in quantitative MRI but lacks robust uncertainty quantification. We provide an efficient, validated framework to generate confidence and/or credible intervals, enabling statistically robust and interpretable estimates essential for clinical confidence.