Using AI to Diagnose and Assess Parkinson's Disease: Challenges, Algorithms, and Applications


Bio: Yuzhe Yang is a PhD student in computer science at MIT CSAIL. He received his B.S. degree in EECS at Peking University. His research interests include machine learning and AI for healthcare. His research focuses on fundamental machine learning algorithms for model robustness and generalization to enable real-world applications especially in health domain, as well as building innovative learning systems to enable new modalities and frameworks for digital health. His research has been published at top interdisciplinary journals and AI/ML conferences including Nature Medicine, Science Translational Medicine, NeurIPS, ICML, ICLR, CVPR, and ECCV, and his works have been recognized by the MathWorks Fellowship, and media coverage from MIT Tech Review, Wall Street Journal, Forbes, BBC, The Washington Post, etc.



There are currently no effective biomarkers for diagnosing Parkinson’s disease (PD) or tracking its progression. In this talk, I will present an artificial intelligence (AI) model to detect PD and track its progression from nocturnal breathing signals [1]. I’ll first discuss background, problem setup, and general challenges and principles for designing such AI models in the wild for health applications: sparse supervision, data imbalance, and distribution shift. I’ll present the most general form of each principle before providing concrete instantiations of using each in practice. This will include a simple multi-task learning method that incorporates health domain knowledge and for AI model interpretation, a framework for learning imbalanced data with continuous targets [2], and an algorithm that enables learning from multi-domain imbalanced data as well as imbalanced domain generalization with theoretical guarantees [3]. Finally, I will conclude with implications and applications of using AI to advance digital medicine and other real-world applications.


  1. Artificial intelligence-enabled detection and assessment of Parkinson’s disease using nocturnal breathing signals.
  2. Delving into Deep Imbalanced Regression.
  3. On Multi-Domain Long-Tailed Recognition, Imbalanced Domain Generalization and Beyond.