TY - JOUR AU - Krueger, Felix AU - Aigner, Christoph Stefan AU - Lutz, Max AU - Riemann, Layla Tabea AU - Degenhardt, Katja AU - Hadjikiriakos, Kimon AU - Zimmermann, Felix Frederik AU - Hammernik, Kerstin AU - Schulz-Menger, Jeanette AU - Schaeffter, Tobias AU - Schmitter, Sebastian TI - Deep learning-based whole-brain B1 +-mapping at 7T. JO - Magnetic resonance in medicine VL - 93 IS - 4 SN - 1522-2594 CY - New York, NY [u.a.] PB - Wiley-Liss M1 - DKFZ-2024-02171 SP - 1700-1711 PY - 2025 N1 - #LA:E020# / 2025 Apr;93(4):1700-1711 AB - This study investigates the feasibility of using complex-valued neural networks (NNs) to estimate quantitative transmit magnetic RF field (B1 +) maps from multi-slice localizer scans with different slice orientations in the human head at 7T, aiming to accelerate subject-specific B1 +-calibration using parallel transmission (pTx).Datasets containing channel-wise B1 +-maps and corresponding multi-slice localizers were acquired in axial, sagittal, and coronal orientation in 15 healthy subjects utilizing an eight-channel pTx transceiver head coil. Training included five-fold cross-validation for four network configurations: NN cx tra AB - <br clear="all" /><table border="0" width="100%"><tr><td> AB - <table align="center" cellspacing="0" cellpadding="2"><tr><td nowrap="nowrap" align="center"> AB - NN<sub>cx</sub><sup>tra</sup> </td></tr></table> AB - </td></tr></table> AB - AB - used transversal, NN cx sag AB - <br clear="all" /><table border="0" width="100%"><tr><td> AB - <table align="center" cellspacing="0" cellpadding="2"><tr><td nowrap="nowrap" align="center"> AB - NN<sub>cx</sub><sup>sag</sup> </td></tr></table> AB - </td></tr></table> AB - AB - sagittal, NN cx cor AB - <br clear="all" /><table border="0" width="100%"><tr><td> AB - <table align="center" cellspacing="0" cellpadding="2"><tr><td nowrap="nowrap" align="center"> AB - NN<sub>cx</sub><sup>cor</sup> </td></tr></table> AB - </td></tr></table> AB - AB - coronal data, and NN cx all AB - <br clear="all" /><table border="0" width="100%"><tr><td> AB - <table align="center" cellspacing="0" cellpadding="2"><tr><td nowrap="nowrap" align="center"> AB - NN<sub>cx</sub><sup>all</sup> </td></tr></table> AB - </td></tr></table> AB - AB - was trained on all slice orientations. The resulting maps were compared to B1 +-reference scans using different quality metrics. The proposed network was applied in-vivo at 7T in two unseen test subjects using dynamic kt-point pulses.Predicted B1 +-maps demonstrated a high similarity with measured B1 +-maps across multiple orientations. The estimation matched the reference with a mean relative error in the magnitude of (2.70 ± 2.86) KW - 7 tesla (Other) KW - B1+‐mapping (Other) KW - brain (Other) KW - deep learning (Other) KW - parallel transmission (Other) LB - PUB:(DE-HGF)16 C6 - pmid:39462473 DO - DOI:10.1002/mrm.30359 UR - https://inrepo02.dkfz.de/record/294338 ER -