The director of Laryngology Research at Mass Eye and Ear spoke to the use of machine learning programs to aid physicians in the diagnosis of dystonia.
“Clinicians just don’t have the luxury of several hours per patient to process their data. In a way, our study was designed to lessen the time burden on the clinician and to speed up the diagnosis.”
Recently, Kristina Simonyan, MD, PhD, DrMed, and Davide Valeriani, PhD, published the results of an assessment of artificial intelligence (AI) based deep learning platform, dubbed DystoniaNet, which uses magnetic resonance imaging (MRI) data to identify patients with focal dystonia. The results of this data-driven approach revealed that DystoniaNet could diagnose the disorder with near 100% accuracy in less than a second.
Previously, Simonyan and Valeriani had conducted an analysis-driven assessment to test this method, and the results were encouraging, offering an accuracy between 71% and 81% of correctly diagnosing dystonia. Additionally, similar research from other laboratories has shown analysis-driven biomarkers can be implemented for diagnosing dystonia. But for Simonyan, that was not quite good enough. That type of paradigm, she told NeurologyLive, involves far too much data processing for the physician, which is clinically challenging.
Simonyan, who is the director of Laryngology Research at Mass Eye and Ear, as well as an associate neuroscientist at Massachusetts General Hospital and associate professor of Otolaryngology-Head and Neck Surgery at Harvard Medical School, offered her perspective on the use of machine learning and AI tools like DystoniaNet in this conversation with NeurologyLive.