2026-05-19T05:07:59Zhttps://keep.lib.asu.edu/oai/request

oai:keep.lib.asu.edu:node-2014002025-05-09T20:34:29Zoai_pmh:repo_items

201400 https://hdl.handle.net/2286/R.2.N.201400 http://rightsstatements.org/vocab/InC/1.0/ http://creativecommons.org/licenses/by-nc-sa/4.0 2025-05 2027-05-09T05:00:00 41 pages Puvvadi, Suraj Ghasemzadeh, Hassan Peterson, Daniel Barua Soumma, Shovito Barrett, The Honors College College of Health Solutions Parkinson’s disease is a progressive, neurodegenerative disorder affecting one million Americans, with an annual economic burden of $52 billion. Early symptoms often go unnoticed or are mistaken for other movement disorders, delaying final clinical diagnosis by 2.75 years. We propose a transformer model using attention mechanisms with hyperparameter tuning to analyze sensor data and reveal imperceptible temporal patterns. We leverage embedded systems to improve real-time tracking, reduce diagnostic delays, and enable early interventions. Deep learning Parkinson's disease Smartwatch Sensors Transformer Deep Learning Approaches for Classifying Parkinson’s Disease Using Smartwatch-Based Physiological Signals