DystoniaDBSNet as a novel deep learning biomarker of predictive deep brain stimulation outcome in dystonia

Dystonia is a debilitating movement disorder that interferes with daily activities and significantly impacts patients’ quality of life. Among the therapeutic options is deep brain stimulation of globus pallidus internus (GPi-DBS); however, there are no objective markers or standard tests for pre-surgical candidate selection or efficacy assessment. Using brain MRIs of 175 dystonia patients from five international clinical centers, we developed (n = 104), tested (n = 30), and externally validated (n = 41) a deep learning algorithm, DystoniaDBSNet, to objectively determine GPi-DBS response based on its automatically discovered neural marker of treatment outcome.

DystoniaDBSNet achieved an overall accuracy of 95.5%, with 98.0% sensitivity, 87.5% specificity, and a 5.6% referral rate. The average computational time was 36.2 seconds per case. The algorithmic decision was based on its neural marker comprising clusters in premotor cortex, primary sensorimotor cortex, supplementary motor area, parietal lobule, thalamus, inferior fronto-occipital fasciculus, and corpus callosum.

DystoniaDBSNet provides a fully automated, objective, accurate, and fast predictive assessment of GPi-DBS outcome in patients with different forms of isolated dystonia, based on structural brain MRI. As such, DystoniaDBSNet may offer a data-driven approach to evaluating GPi-DBS candidacy of dystonia patients, which, in turn, could help increase its utilization in this population with limited therapeutic options. We acknowledge the use of the Medical Imaging Informatics Bench to Bedside (mi2b2) workbench for image retrieval.

We thank Davide Valeriani, PhD, for initial assistance with this study. This study was funded by the National Institute of Neurological Disorders and Stroke and National Institute on Deafness and Other Communication Disorders, National Institutes of Health (R01NS124228, R01DC011805, P50DC019900 to KS and P01NS087997 to NS) and the Amazon Web Services Machine Learning Research Award to K.S. The funders had no role in this study.

Department of Otolaryngology-Head and Neck Surgery, Massachusetts Eye and Ear and Harvard Medical School, Boston, MA, USA Dongren Yao, Giovanni Battistella & Kristina Simonyan Department of Neurology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA Department of Neurological Surgery, Albert Einstein College of Medicine – Montefiore Medical Center, Bronx, NY, USA Department of Clinical and Movement Neurosciences, UCL Queen Square Institute of Neurology, London, UK Eoin Mulroy, Kailash Bhatia, Thomas Foltynie, Patricia Limousin, Harith Akram & Ludvic Zrinzo Department of Neurology, University of Florida, Gainesville, FL, USA Department of Neurology, University of Pennsylvania, Philadelphia, PA, USA Andres Deik, Gordon H. Baltuch & Meredith Spindler Department of Neurosurgery, Columbia University Irving Medical Center, New York, NY, USA Department of Neurosurgery, Tokyo Women’s Medical University, Tokyo, Japan G.B., N.B., E.E., E.M., K.B., J.K.W., J.J., T.F., P.L., H.A., L.Z., G.H.B., M.S., K.K., S.H. declare no competing financial or non-financial interests. A.D. reports receiving personal compensation for consulting services for Amneal Pharmaceuticals, Manifold Bio, Genus Lifesciences, and SPARK-NS, serving on Scientific Advisory or Data Safety Monitoring Boards for Supernus Therapeutics and Amneal Pharmaceuticals, receiving research support from Teva Pharmaceuticals, Cerevel Therapeutics, Prevail Therapeutics, Lundbeck Therapeutics, AskBio, Annovis Bio, Ono Pharmaceuticals, Amylyx Therapeutics, and the Dystonia Coalition/Dystonia Medical Research Foundation, and receiving publishing royalties related to health care publications with UpToDate and Elsevier, all unrelated to the present work.

K.S. reports receiving funding from the National Institutes of Health (R01NS088160, R01NS124228, R01DC011805, R01DC012545, R01DC019353, P50DC01990), the Department of Defense, and Amazon Web Services, and serving on the Scientific Advisory Board of the Tourette Association of America and The Voice Foundation. K.S. is an inventor on a granted United States Patent (US 12,257,063) and pending European Patent (EP20871026.9) relevant to this work, and an inventor on pending United States Patents (US 62/908,442, US 63/910,006) unrelated to this work. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.