Model Generalizability Investigation for GFCE-MRI Synthesis in
Radiotherapy of NPC patients Using Multi-institutional Data and
Patient-based Data Normalization
Abstract
In this study, we aimed at investigating generalizability of GFCE-MRI
model using data from seven institutions by manipulating heterogeneity
of training MRI data under two popular normalization approaches. A
multimodality-guided synergistic neural network (MMgSN-Net) was applied
to map from T1-weighted and T2-weighted MRI to contrast-enhanced MRI
(CE-MRI) for GFCE-MRI synthesis in patients with nasopharyngeal
carcinoma. MRI data from three institutions were used separately to
generate three uni-institution models and jointly for a tri-institution
model. Min-Max and Z-Score were applied for data normalization of each
model. MRI data from the remaining four institutions served as external
cohorts for model generalizability assessment. Quality of GFCE-MRI was
quantitatively evaluated against ground-truth CE-MRI using mean absolute
error (MAE) and peak signal-to-noise ratio (PSNR). Results showed that
performance of all uni-institution models remarkably dropped on the
external cohorts. By contrast, model trained using multi-institutional
data with Z-Score normalization yielded significantly improved model
generalizability