Magnetic Resonance Fingerprinting with Combined Gradient- and Spin-echo Echo-planar Imaging: Simultaneous Estimation of T1, T2 and T2* with integrated-B1 Correction

Abstract

PurposeTo introduce a novel magnetic-resonance fingerprinting (MRF) framework with single-shot echo-planar imaging (EPI) readout to simultaneously estimate tissue T2, T1 and T2*, and integrate B1 correction.\n\nMethodsSpin-echo EPI is combined with gradient-echo EPI to achieve T2 estimation as well as T1 and T2* quantification. In the dictionary matching step, the GE-EPI data segment provides estimates of tissue T1 and T2* with additional B1 information, which are then incorporated into the T2-matching step that uses the SE-EPI data segment. In this way, biases in T2 and T2* estimates do not affect each other.\n\nResultsAn excellent correspondence was found between our T1, T2, and T2* estimates and results obtained from standard approaches in both phantom and human scans. In the phantom scan, a linear relationship with R2>0.96 was found for all parameter estimates. The maximum error in the T2 estimate was found to be below 6%. In the in-vivo scan, similar contrast was noted between MRF and standard approaches, and values found in a small region of interest (ROI) located in the grey matter (GM) were in line with previous measurements (T2MRF=88{+/-}7ms vs T2Ref=89{+/-}11ms, T1MRF=1153{+/-}154ms vs T1Ref=1122{+/-}52ms, T2*MRF=56{+/-}4ms vs T2*Ref=53{+/-}3ms).\n\nConclusionAdding a spin echo data segment to EPI based MRF allows accurate and robust measurements of T2, T1 and T2* relaxation times. This MRF framework is easier to implement than spiral-based MRF. It doesnt suffer from undersampling artifacts and seems to require a smaller dictionary size that can fasten the reconstruction process.