Lesion probability mapping in MS patients using a regression network on MR fingerprinting

Ingo Hermann*, Alena K. Golla, Eloy Martínez-Heras, Ralf Schmidt, Elisabeth Solana, Sara Llufriu, Achim Gass, Lothar R. Schad, Frank G. Zöllner

*Corresponding author for this work

Research output: Contribution to journalArticleScientificpeer-review

15 Downloads (Pure)


Background: To develop a regression neural network for the reconstruction of lesion probability maps on Magnetic Resonance Fingerprinting using echo-planar imaging (MRF-EPI) in addition to T1, T2∗, NAWM, and GM- probability maps. Methods: We performed MRF-EPI measurements in 42 patients with multiple sclerosis and 6 healthy volunteers along two sites. A U-net was trained to reconstruct the denoised and distortion corrected T1 and T2∗ maps, and to additionally generate NAWM-, GM-, and WM lesion probability maps. Results: WM lesions were predicted with a dice coefficient of 0.61 ± 0.09 and a lesion detection rate of 0.85 ± 0.25 for a threshold of 33%. The network jointly enabled accurate T1 and T2∗ times with relative deviations of 5.2% and 5.1% and average dice coefficients of 0.92 ± 0.04 and 0.91 ± 0.03 for NAWM and GM after binarizing with a threshold of 80%. Conclusion: DL is a promising tool for the prediction of lesion probability maps in a fraction of time. These might be of clinical interest for the WM lesion analysis in MS patients.

Original languageEnglish
Article number107
Number of pages11
JournalBMC Medical Imaging
Issue number1
Publication statusPublished - 2021


  • Deep learning reconstruction
  • Lesion prediction
  • Magnetic resonance fingerprinting
  • T Mapping
  • T* Mapping


Dive into the research topics of 'Lesion probability mapping in MS patients using a regression network on MR fingerprinting'. Together they form a unique fingerprint.

Cite this