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.

Purpose: To evaluate the use of magnetic resonance fingerprinting (MRF) for simultaneous quantification of T₁ and T*₂ in a single breath-hold in the kidneys. Methods: The proposed kidney MRF sequence was based on MRF echo-planar imaging. Thirty-five measurements per slice and overall 4 slices were measured in 15.4 seconds. Group matching was performed for in-line quantification of T₁ and T*₂. Images were acquired in a phantom and 8 healthy volunteers in coronal orientation. To evaluate our approach, region of interests were drawn in the kidneys to calculate mean values and standard deviations of the T₁ and T*₂ times. Precision was calculated across multiple repeated MRF scans. Gaussian filtering is applied on baseline images to improve SNR and match stability. Results: T₁ and T*₂ times acquired with MRF in the phantom showed good agreement with reference measurements and conventional mapping methods with deviations of less than 5% for T₁ and less than 10% for T*₂. Baseline images in vivo were free of artifacts and relaxation times yielded good agreement with conventional methods and literature (deviation T₁:7 ± 4%, T*₂:6±3%). Conclusions: In this feasibility study, the proposed renal MRF sequence resulted in accurate T₁ and T*₂ quantification in a single breath-hold.