3. Leveraging GANs for Super-resolution of Ultra-low-field Paediatric MRI

COEX 

Magnetic resonance imaging is integral for assessment of paediatric neurodevelopment, however modern MRI systems are large and expensive. Recent ultra-low-field MRI systems such as the 64mT Hyperfine Swoop show great promise in widening accessibility to MRI in low-income settings. Imaging at such low field strengths comes at the cost of lower spatial resolution and signal-to-noise ratio, although these can be mitigated via deep-learning super-resolution. Here we investigate the effectiveness of conditional general adversarial networks (GANs) in enhancing the resolution of ultra-low-field images acquired from a paediatric population.