Related Experiment Video
Updated: Jul 6, 2026
Magnetic Resonance Imaging of Multiple Sclerosis at 7.0 Tesla
Published on: February 19, 2021
Accelerated Deep-Learning-Based Image Reconstruction for 3D T2 Dark-Fluid in Imaging of Multiple Sclerosis.
Sebastian Altmann1, Mario A A Mercado, Arwed Elias Michael
1Department of Neuroradiology, University Medical Center Mainz, Johannes Gutenberg University, Langenbeckstr (S.A., M.A.A.M., A.E.M., M.K., V.I.S., M.A.B., A.E.O.); Institute of Medical Biostatistics, Epidemiology and Informatics (IMBEI), University Medical Center Mainz, Johannes Gutenberg University, Rhabanusstr, Mainz (R.H.P.); Research & Clinical Translation, Magnetic Resonance, Siemens Healthineers AG, Erlangen (M.D.N.); Department of Neurology, University Medical Center Mainz, Johannes Gutenberg University, Langenbeckstr, Mainz, Germany (V.F.).
Deep-learning (DL)-accelerated MRI is interchangeable with conventional MRI for detecting new white matter lesions in multiple sclerosis (MS) patients. This advancement significantly cuts scan times by nearly 50%, improving efficiency in MS surveillance.
Area of Science:
- Radiology
- Medical Imaging
- Artificial Intelligence in Medicine
Background:
- Deep-learning (DL) MRI accelerates image acquisition, but its interchangeability with conventional 3D MRI for monitoring multiple sclerosis (MS) requires evaluation.
- Detecting new white matter lesions is crucial for assessing MS disease activity.
Purpose of the Study:
- To investigate the interchangeability and comparability of conventional 3D-T2-SPACE dark-fluid (c-3D-T2) and DL-accelerated 3D-T2-SPACE dark-fluid (DL-3D-T2) sequences.
- To assess the efficacy of DL-accelerated MRI in detecting new white matter lesions in patients with MS or suspected chronic inflammatory CNS disease.
Main Methods:
- Prospective study of 94 patients (77 MS, 17 CNS disease) using 1.5T brain MRI.
- Comparison of c-3D-T2 (5:01 min) and DL-3D-T2 (2:48 min) sequences for detecting new white matter lesions.
- Interchangeability assessed using the 2024 revised McDonald criteria, with an equivalence margin of 5% and Gwet's AC1/AC2 for inter-reader reliability.
Main Results:
- DL-3D-T2 demonstrated interchangeability with c-3D-T2 for detecting new white matter lesions across all readers.
- Near-perfect agreement (Gwet's AC2 ≥ 0.85) was observed for lesion detection between the two methods.
- While DL-accelerated imaging reduced acquisition time by nearly 50%, subjective image quality and diagnostic confidence were rated slightly superior for conventional imaging by some readers.
Conclusions:
- DL-accelerated 3D-T2-weighted imaging is interchangeable with conventional 3D imaging for detecting new white matter lesions in MS.
- The significant reduction in acquisition time supports more efficient MRI protocols for MS surveillance.
- Optimizing DL acceleration factors may be considered to balance time savings with enhanced subjective image quality.
