Multimodal AI for Parkinson's Disease

University of Sheffield Collaborating Faculties: Sheffield Institute for Translational Neuroscience (SITraN) and Faculty of Engineering

Overview: Our project is dedicated to the development of artificial intelligence tools for Parkinson’s Disease, designed to elucidate the underlying mechanisms of the condition and predict its progression. This initiative integrates data from multiple modalities—including genetic data, biomarkers, environmental factors, and medical examinations—utilizing advanced AI methodologies such as contrastive learning, foundation models, and causal discovery.

Motivation: The growing interest in applying artificial intelligence (AI) to tackle Parkinson’s Disease reflects a comprehensive appreciation of the condition’s widespread impact, its escalating incidence, the existing gaps in our comprehension, and the extraordinary research opportunities afforded by current data repositories.

As the second most prevalent neurodegenerative disorder in the UK, Parkinson’s Disease presents a formidable public health challenge. Its widespread nature underlines the pressing demand for novel treatment and management strategies, positioning AI as an ideal candidate to drive forward innovative solutions. Globally, the frequency of Parkinson’s Disease is on the rise, affecting an ever-growing number of individuals either directly or putting them at a significant risk of developing the condition. This increasing trend accentuates the urgent necessity for interventions that are both scalable and efficacious, areas where AI technology shines with potential. Despite thorough research efforts, the core mechanisms behind Parkinson’s Disease remain a mystery. Here, AI’s ability to sift through and analyze large, complex datasets could unlock new understanding, setting the stage for transformative developments in how we treat and prevent the disease.

The UK Biobank, with its comprehensive data, emerges as a pivotal resource for AI research. The extensive scope of this data lays a robust groundwork for the creation and refinement of AI models, fostering considerable progress in our grasp and handling of Parkinson’s. Leveraging AI, researchers are not just shedding light on the elusive causes of Parkinson’s Disease but are also crafting predictive models that foresee the disease’s trajectory, pinpoint therapeutic targets, and ultimately, instill hope in the countless lives touched by this afflictive illness.

Thus, the convergence of AI and Parkinson’s Disease research heralds an exciting era in the battle against neurodegenerative diseases. It offers a beacon of hope for enhancing patient outcomes, marking a vital step forward in our journey towards understanding, managing, and eventually overcoming such conditions.

Wenrui Fan
Wenrui Fan
AI Research Engineer
Thomas W. Payne
Thomas W. Payne
NIHR Clinical Lecturer in Neurology
Mohammod Suvon
Mohammod Suvon
AI Research Engineer
Xianyuan Liu
Xianyuan Liu
Assistant Head of AI Research Engineering, and Senior AI Research Engineer
Haolin Wang
Haolin Wang
AI Research Engineer
Jiayang Zhang
Jiayang Zhang
Deputy Assistant Head of AI Research Engineering, and AI Research Engineer
Venet Osmani
Venet Osmani
Senior Lecturer in Data Science
Chen Chen
Chen Chen
Lecturer in Computer Vision
Shuo Zhou
Shuo Zhou
Deputy Head of AI Research Engineering, and Academic Fellow in Machine Learning
Oliver Bandmann
Oliver Bandmann
Professor of Movement Disorders Neurology, Honorary Consultant Neurologist, and Co-Director of Neuroscience Institute
Haiping Lu
Haiping Lu
Head of AI Research Engineering, Professor of Machine Learning, and Turing Academic Lead