Longitudinal Prediction of Postnatal Brain Magnetic Resonance Images via a Metamorphic Generative Adversarial Network

TL;DR

This paper introduces MGAN, a deep learning model that predicts missing infant brain MRI scans over time, effectively handling rapid developmental changes and improving the accuracy of longitudinal studies.

Contribution

The paper presents a novel metamorphic GAN architecture with spatial-frequency translation, quality-guided learning, and multi-scale loss for accurate MRI prediction in infant studies.

Findings

MGAN outperforms existing GANs in MRI translation accuracy.

It effectively captures contrast and structural details during rapid brain development.

The model enhances longitudinal MRI analysis in infant research.

Abstract

Missing scans are inevitable in longitudinal studies due to either subject dropouts or failed scans. In this paper, we propose a deep learning framework to predict missing scans from acquired scans, catering to longitudinal infant studies. Prediction of infant brain MRI is challenging owing to the rapid contrast and structural changes particularly during the first year of life. We introduce a trustworthy metamorphic generative adversarial network (MGAN) for translating infant brain MRI from one time-point to another. MGAN has three key features: (i) Image translation leveraging spatial and frequency information for detail-preserving mapping; (ii) Quality-guided learning strategy that focuses attention on challenging regions. (iii) Multi-scale hybrid loss function that improves translation of tissue contrast and structural details. Experimental results indicate that MGAN outperforms existing GANs by accurately predicting both contrast and anatomical details.