Rapid multi-parametric quantitative MRI via deep learning-based synthetic-to-real reconstruction and 3D SSFP-MOLED imaging

Clinical Snapshot

Key Findings

• P Value: Not reported

• Effect Size: Not quantitatively reported in the abstract; qualitative claims of 'high accuracy and reproducibility' without numerical effect estimates

• Primary Outcome: Simultaneous whole-brain quantification of six MRI parameters (M0, T1, T2, T2*, B1+, ΔB0) at 1×1×2 mm³ isotropic resolution within approximately 3 minutes using 2× parallel imaging acceleration, with reported high accuracy and reproducibility in phantom, healthy volunteer, and clinical tumour/haemorrhage settings

• Nnt Or Sensitivity: Not applicable at this stage of technical validation; no diagnostic accuracy metrics (sensitivity, specificity, AUC) reported. Scan time reduction from conventional multi-sequence qMRI protocols (typically 20–60 minutes for equivalent parameter coverage) to approximately 3 minutes represents the primary efficiency metric, though direct comparative timing data are not provided in the abstract.

• Confidence Interval: Not reported

Clinical Application

Currently a research-stage technique requiring specialised pulse sequence implementation, physics-constrained synthetic data generation infrastructure, and trained deep learning reconstruction networks. Clinical deployment would require regulatory approval, site-specific validation, radiologist training, and integration with PACS/RIS workflows. The 2× parallel imaging acceleration requirement is compatible with most modern clinical 3T MRI scanners. Scalability to 1.5T systems and non-brain applications has not been demonstrated. Australia has no current TGA-approved indication specifically for 3D SSFP-MOLED or equivalent multi-parametric qMRI techniques. Quantitative MRI is not currently reimbursed under the Medicare Benefits Schedule (MBS) as a distinct item — parametric maps are not billable separately from standard MRI sequences. RACGP and RANZCR guidelines do not yet incorporate multi-parametric qMRI recommendations for routine brain imaging. If validated in larger prospective studies, this technology could support RANZCR advocacy for MBS reform to include quantitative neuroimaging. Australian academic MRI centres (e.g., Melbourne Brain Centre, Florey Institute, NeuRA) would be appropriate sites for independent replication studies. The technique's field-inhomogeneity robustness may be particularly relevant in regional Australian centres where scanner maintenance standards vary. PBS implications are not directly relevant as this is an imaging rather than pharmaceutical intervention. Adult patients requiring brain MRI with quantitative tissue characterisation, particularly those with suspected or confirmed intracranial neoplasms, haemorrhage, demyelinating disease, or neurodegenerative conditions where multi-parametric tissue mapping adds diagnostic value. Paediatric and motion-prone populations may particularly benefit from reduced scan times.

Abstract

Multi-parametric quantitative magnetic resonance imaging (mqMRI) holds significant clinical potential through multi-parametric tissue characterization, yet its adoption is hindered by prolonged scan time and sensitivity to non-ideal signal conditions, especially in high-resolution whole-brain protocols. To address these challenges, we propose a novel signal encoding method integrating phase-modulated three dimensional steady-state free precession with multiple overlapping-echo detachment (3D SSFP-MOLED). This method simultaneously encodes six physiological parameters (M0, T1, T2, T2*, B1+, ΔB0) into k-space by controlling overlapping echo detachment in signal acquisition. A physics-constrained synthetic data pipeline was developed to simulate MR signal evolutions with realistic field variations (ΔB0, B1+ inhomogeneities), enabling robust training of network for real-time parameter mapping. Whole-brain parametric maps (1×1×2 mm³ resolution) can be delivered within 3 minutes with only 2x parallel acquisition acceleration. Validation was performed on phantom, healthy volunteers, and clinical cases with tumors/hemorrhage. Experimental results show that our method can achieve rapid multi-parametric quantitation with high accuracy and reproducibility. By synergizing adaptive signal encoding, physics-informed synthetic training, and reproducible deep learning reconstruction, this work establishes a new paradigm for efficient and reliable mqMRI in clinical signal processing applications.

References

1. 1.Yang, J., Zhu, L., Xiong, K., Bao, J., Yang, Q., Chen, W., Kang, T., Zhou, J., Lin, J., Lin, L., Chen, Z., Cai, S., & Cai, C. (2026). Rapid multi-parametric quantitative MRI via deep learning-based synthetic-to-real reconstruction and 3D SSFP-MOLED imaging. NeuroImage. https://doi.org/10.1016/j.neuroimage.2026.121985