Learning An Mr Acquisition-Invariant Representation Using Siamese Neural Networks

Generalization of voxelwise classifiers is hampered by differences between MRI-scanners, e.g. different acquisition protocols and field strengths. To address this limitation, we propose a Siamese neural network (MRAI-NET) that extracts acquisition-invariant feature vectors. These can consequently be used by task-specific methods, such as voxelwise classifiers for tissue segmentation. MRAI-NET is evaluated on both simulated and real patient data. Experiments show that MRAI-NET outperforms both voxelwise classifiers trained on the source data as well as classifiers trained on the limited amount of target scanner data available.