Protocol harmonization using a generative adversarial network decreases morphometry variability
Veronica Ravano, Jean‐François Démonet, Daniel Damian, Reto Antoine Meuli, Gian Franco Piredda, Till Huelnhagen, Bénédicte Maréchal, Jean‐Philippe Thiran, Tobias Kober, Jonas Richiardi · Proceedings on CD-ROM - International Society for Magnetic Resonance in Medicine. Scientific Meeting and Exhibition/Proceedings of the International Society for Magnetic Resonance in Medicine, Scientific Meeting and Exhibition · 2023
In radiology, the deployment of automated clinical decision support tools to new institutions is often hindered by inter-site data variability. In MRI, data heterogeneity often arises from differences in acquisition protocols. To overcome this issue, we propose a post-hoc harmonization technique based on generative adversarial networks (GAN). Seventy-seven patients suffering from dementia were scanned with two distinct T1-weighted MP-RAGE protocols. We show that cross-protocol harmonization of brain images using a conditional GAN improves image similarity and reduces the variability of brain morphometry.