Tariq Adnan — AI in Healthcare & Multimodal Learning
I am a PhD candidate in Computer Science at the University of Rochester (ROC-HCI Lab), where my research focuses on multimodal learning for AI-driven healthcare. I develop scalable frameworks for automated screening and personalized management of Parkinson’s disease (PD), using video and audio analysis to enable remote, accessible assessments that lower barriers to specialty care.
My current work centers on three directions:
- Remote PD screening: detecting Parkinson’s disease using webcam-based facial, speech, and motor assessments.
- Longitudinal digital biomarkers: tracking symptom severity over time from standardized at-home recordings.
- LLM-powered PD assistants: contextualizing severity trends, medication timing, and lifestyle factors to support individualized PD management.
Before starting my PhD, I served as an Assistant Professor in the Department of Computer Science and Engineering at BUET, working at the intersection of distributed systems, big data analytics, and applied machine learning. Across these projects, my long-term goal is to translate AI methods into reliable, user-centered tools that meaningfully support patients, clinicians, and healthcare systems.
Recent Highlights
- Dec 2025: Invited to present our NEJM AI work at NYU Langone Health (March 2026).
- Oct 2025: Hi5: Synthetic Data for Hand Pose Estimation accepted at ACII 2025.
- Jun 2025: AI-Enabled PD Screening Using Smile Videos published in NEJM AI.
- Jun 2025: Fusion Architecture for PD Detection published in npj Parkinson’s Disease.
- Feb 2025: Accessible, At-Home Detection of PD published in AAAI 2025.
- Jan 2024: User-Centered Framework for PD published in IMWUT.
- Dec 2023: Received Best Abstract Award (Clinical) at the Future of Parkinson’s Conference.
- Dec 2022: UACD: Influential Spreaders in Twitter published in SNAM.
- Sep 2022: Started PhD at University of Rochester (ROC-HCI Lab).
- Jun 2021: Geo-Distributed PCA published in Information Systems.