Antoine Moevus1,2, Mathieu Dehaes2,3,4, and Benjamin De Leener1,2,4
1Departement of Computer and Software Engineering, Polytechnique Montréal, Montréal, QC, Canada, 2Institute of Biomedical Engineering, University of Montreal, Montréal, QC, Canada, 3Department of Radiology, Radio-oncology, and Nuclear Medicine, Université de Montréal, Montréal, QC, Canada, 4Research Center, Ste-Justine Hospital University Centre, Montreal, QC, Canada
1Departement of Computer and Software Engineering, Polytechnique Montréal, Montréal, QC, Canada, 2Institute of Biomedical Engineering, University of Montreal, Montréal, QC, Canada, 3Department of Radiology, Radio-oncology, and Nuclear Medicine, Université de Montréal, Montréal, QC, Canada, 4Research Center, Ste-Justine Hospital University Centre, Montreal, QC, Canada
We explored the application of deep learning for superresolution on QSM data. We demonstrated the importance of training strategies for superresolution. We evaluated different loss functions for training neural networks on brain QSM data.
Figure 3 - Qualitative comparison of the models on a selected slice for the whole brain and two selected regions (green and blue boxes). The AHEAD reference is subject 005 slice 158. The ** represents the unoptimized cyclic LR WIMSE.
Table 1 - Comparison of models performance using the MSE and SSIM metrics. MSE: low score refers to high performance, SSIM: high score refers to high performance. Best performances are highlighted.