Rahul Gaurav, PhD, Paris Brain Institute, Paris, France, describes the integration of deep-learning artificial intelligence models in processing neuromelanin-sensitive MRI (NM-MRI) images. NM-MRI is a leading method for analyzing spatiotemporal physiological changes in brain regions in neurodegenerative disease. Currently, images are segmented into regions for analysis manually, meaning the quality and consistency of research is highly dependent on the experience of the experimenter involved. Dr Gaurav and colleagues are currently developing a convolutional neural network (ConvNet)-based architecture called the U-net. This model returned 80% similarity to manually performed segmentation as part of the ongoing ICEBERG study. In the future, it is hoped that such models will return greater accuracy and expand their capability to recognize a range of MRI cohorts and machine types. This interview was recorded at the World Sleep Congress 2022 in Rome, Italy.