From Signals to Intelligence: Hands-on AI in Functional Neuroimaging for Parkinson's Disease
Background and Rationale
Advances in functional neuroimaging and artificial intelligence (AI) are rapidly transforming the field of clinical neuroscience. Techniques such as functional MRI (fMRI) provide rich, high-dimensional data that hold great potential for improving diagnosis, monitoring disease progression, and understanding pathophysiological mechanisms in neurological disorders such as Parkinson's disease (PD).
Despite these advances, a substantial gap remains between data acquisition and practical clinical or research implementation. Many clinicians and early-career researchers lack hands-on experience in processing neuroimaging data, extracting meaningful features, and applying AI-based analytical methods. Furthermore, the recent emergence of generative AI (GenAI) offers new opportunities for data augmentation and model development yet remains largely inaccessible to non-programmers.
This workshop is designed to bridge this gap by providing a structured, hands-on learning experience that spans the full analytical pipeline—from fMRI preprocessing to machine learning and generative AI applications—within a clinically relevant PD framework.
Overview
This hands-on workshop introduces practical AI applications in functional imaging and biomedical signal analysis for Parkinson's disease. Participants will learn how to process fMRI data, transform imaging-derived matrices into machine-learning models, and use generative AI-assisted coding to analyze sleep signals for identifying REM sleep behavior disorder as a prodromal marker of Parkinson's disease.
Learning Objectives
By the end of this workshop, participants will be able to:
- Understand the fundamental workflow of fMRI preprocessing using SPM
- Extract useful features from fMRI-derived matrices.
- Build simple machine-learning models for Parkinson's disease cognition.
- Understand basic 1-D sleep signal processing.
- Use generative AI-assisted "vibe coding" for signal analysis.
- Explore RBD identification as a prodromal marker of Parkinson's disease.
Target Audience
Clinicians, clinical neurophysiologists, sleep specialists, movement disorder researchers, trainees, and biomedical AI beginners interested in applying imaging and signal data to neurological disorders.
- Neurologists and clinical neurophysiologists
- Neuroscience researchers and trainees
- Early career investigators interested in AI applications
- Participants with minimal or no coding experience
