Within-Modality Synthesis and Novel Radiomic Evaluation of Brain MRI Scans

One of the most common challenges in brain MRI scans is to perform different sequences depending on type and properties tissues. In this paper, we propose a generative method translate T2-Weighted (T2W) Magnetic Resonance Imaging (MRI) volume from T2-weight-Fluid-attenuated-Inversion-Recovery (FLAIR) vice versa using Generative Adversarial Networks (GAN). To evaluate proposed method, novel evaluation schema for synthetic approaches based radiomic features. For purpose, consider 510 pair-slices 102 patients train two GAN-based architectures Cycle GAN Dual Cycle-Consistent network (DC2Anet). The results indicate that methods can produce similar original sequence without significant change radiometric feature. Therefore, such assist clinics make decisions generated image when are not available or there enough time re-perform scans.