Premium
Multi‐parametric artificial neural network fitting of phase‐cycled balanced steady‐state free precession data
Author(s) -
Heule Rahel,
Bause Jonas,
Pusterla Orso,
Scheffler Klaus
Publication year - 2020
Publication title -
magnetic resonance in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.696
H-Index - 225
eISSN - 1522-2594
pISSN - 0740-3194
DOI - 10.1002/mrm.28325
Subject(s) - relaxometry , computer science , phase (matter) , parametric statistics , relaxation (psychology) , voxel , artificial neural network , artificial intelligence , nuclear magnetic resonance , physics , magnetic resonance imaging , mathematics , statistics , neuroscience , spin echo , medicine , quantum mechanics , radiology , biology
Purpose Standard relaxation time quantification using phase‐cycled balanced steady‐state free precession (bSSFP), eg, motion‐insensitive rapid configuration relaxometry (MIRACLE), is subject to a considerable underestimation of tissue T 1 and T 2 due to asymmetric intra‐voxel frequency distributions. In this work, an artificial neural network (ANN) fitting approach is proposed to simultaneously extract accurate reference relaxation times (T 1 , T 2 ) and robust field map estimates ( B 1 + , ΔB 0 ) from the bSSFP profile. Methods Whole‐brain bSSFP data acquired at 3T were used for the training of a feedforward ANN with N = 12, 6, and 4 phase‐cycles. The magnitude and phase of the Fourier transformed complex bSSFP frequency response served as input and the multi‐parametric reference set [T 1 , T 2 , B 1 + , ∆B 0 ] as target. The ANN predicted relaxation times were validated against the target and MIRACLE. Results The ANN prediction of T 1 and T 2 for trained and untrained data agreed well with the reference, even for only four acquired phase‐cycles. In contrast, relaxometry based on 4‐point MIRACLE was prone to severe off‐resonance‐related artifacts. ANN predicted B 1 + and ∆B 0 maps showed the expected spatial inhomogeneity patterns in high agreement with the reference measurements for 12‐point, 6‐point, and 4‐point bSSFP phase‐cycling schemes. Conclusion ANNs show promise to provide accurate brain tissue T 1 and T 2 values as well as reliable field map estimates. Moreover, the bSSFP acquisition can be accelerated by reducing the number of phase‐cycles while still delivering robust T 1 , T 2 , B 1 + , and ∆B 0 estimates.