The director of IT and Neuroinformatics Development at the Buffalo Neuroimaging Analysis Center discussed the potential advantages AI-built neuroimaging brings to neurology.
"We were able to use that data to train an artificial intelligence system to automatically delineate and locate the thalamus on T2 flare image, which is the lowest common denominator of imaging."
By leveraging new advances in machine learning and artificial intelligence (AI), the Buffalo Neuroimaging Center (BNAC) has been able to develop numerous new MRI acquisition and analysis methods that are capable of analyzing large-scale datasets with tens of thousands of scans across multiple time points in thousands of individuals.
The efforts from the center coincide with the direction that next-generation precision medicine is headed. Michael Dwyer, PhD, director, IT and Neuroinformatics Development, BNAC, and assistant professor of neurology and biomedical engineering, University at Buffalo, has been a major driver for this type of innovative thinking. To date, he’s published over 150 scientific papers in peer-reviewed medical, imaging, and brain-related journals and reviewed publications for numerous journals in his field. Most recently, he formed an AI special interest group with a specific focus on applications of deep-learning techniques in neuroimaging.
Dwyer claims there are several capabilities and advantages this niche type of technology can bring to neurology. For example, he notes that they have used AI to translate existing metrics from clinical trials to quality, usable images. In an interview with NeurologyLive, Dwyer discussed the hype surrounding AI in neuroimaging, its ability to bridge untranslatable metrics, and its potential use in predicting disease progression in multiple sclerosis.