Predicting dynamic, motion‐related changes in B 0 field in the brain at a 7T MRI using a subject‐specific fine‐trained U‐net

• Published in: Magnetic resonance in medicine Vol. 91; no. 5; pp. 2044 - 2056
• Main Authors: Motyka, Stanislav, Weiser, Paul, Bachrata, Beata, Hingerl, Lukas, Strasser, Bernhard, Hangel, Gilbert, Niess, Eva, Niess, Fabian, Zaitsev, Maxim, Robinson, Simon Daniel, Langs, Georg, Trattnig, Siegfried, Bogner, Wolfgang
• Format: Journal Article
• Language: English
• Published: United States 01.05.2024
• Subjects: Brain - diagnostic imaging; Humans; Image Processing, Computer-Assisted - methods; Magnetic Resonance Imaging - methods; Motion; Movement; Retrospective Studies; artificial neural network; deep learning; magnetic resonance imaging; U-net; patient movement; motion correction; B0 inhomogeneities
• ISSN: 0740-3194; 1522-2594
• DOI: 10.1002/mrm.29980

Subject movement during the MR examination is inevitable and causes not only image artifacts but also deteriorates the homogeneity of the main magnetic field (B ), which is a prerequisite for high quality data. Thus, characterization of changes to B , for example induced by patient movement, is important for MR applications that are prone to B inhomogeneities. We propose a deep learning based method to predict such changes within the brain from the change of the head position to facilitate retrospective or even real-time correction. A 3D U-net was trained on in vivo gradient-echo brain 7T MRI data. The input consisted of B maps and anatomical images at an initial position, and anatomical images at a different head position (obtained by applying a rigid-body transformation on the initial anatomical image). The output consisted of B maps at the new head positions. We further fine-trained the network weights to each subject by measuring a limited number of head positions of the given subject, and trained the U-net with these data. Our approach was compared to established dynamic B field mapping via interleaved navigators, which suffer from limited spatial resolution and the need for undesirable sequence modifications. Qualitative and quantitative comparison showed similar performance between an interleaved navigator-equivalent method and proposed method. It is feasible to predict B maps from rigid subject movement and, when combined with external tracking hardware, this information could be used to improve the quality of MR acquisitions without the use of navigators.