Hi! I am Sydney Pugh, a postdoctoral researcher in the Artificial Intelligence for Ambulatory Care Innovation (AI-4-AI) Lab at the University of Pennsylvania, where I am advised by Kevin Johnson. I am also a Penn AI Fellow and a PennAITech Innovation Fellow. My research focuses on developing trustworthy, interpretable, and user-centered AI tools that can move from development into real clinical use. I am particularly interested in methods that hold up under real-world data constraints, such as small datasets and limited ground truth, and in designing AI systems around the needs of the clinicians and patients who use them.
Currently, I lead the development of WATCH-SS, an explainable AI tool that screens for cognitive impairment from patient speech. My work on WATCH-SS spans model development and is now moving toward clinical validation and integration into primary care workflows. I also contribute to projects on chatbots for primary care and ethical oversight of clinical multi-agent AI systems.
I earned my Ph.D. in Computer and Information Science (CIS) at the University of Pennsylvania in August 2024, where I was advised by Insup Lee and James Weimer. My dissertation explored weakly-supervised methods for rapid and cost-effective performance evaluation of medical AI systems. I also earned my M.S.E. in CIS at UPenn in 2021. Prior to that, I earned my B.S. in Applied Mathematics and Computer Science from Loyola University Maryland in 2019.
News
- October 23, 2026: Our paper “Multilingual Evaluation of a Large Language Model-Based Primary Care Chatbot,” led by former master’s student Patricia Chen, is now available in JAMIA Open.
- August 25, 2026: Preprint “Detecting Self-Repairs from Spontaneous Speech with Prompt Ablation Across LLMs and Fine-Tuned Encoder” is now available on medRxiv. This work, led by master’s student Rachel Wu, will be presented at the Pacific Symposium on Biocomputing (PSB) 2027.
- August 15, 2026: Preprint “ETHOS: Towards a Modular Ethics Framework for Clinical Multi-Agent Systems” (co-first author) is now available on arXiv. I will present it as a poster at the Pacific Symposium on Biocomputing (PSB) 2027.
- November 5, 2025: I was invited to serve as an Innovation Fellow at the Penn Artificial Intelligence and Technologies Collaboratory for Healthy Aging (PennAITech).
- September 9, 2025: Our paper “WATCH-SS: Developing a Trustworthy and Explainable Modular Framework for Detecting Cognitive Impairment from Spontaneous Speech” was accepted to the Pacific Symposium on Biocomputing (PSB26) and presented as a poster in Big Island, Hawaii.
- August 21, 2025: Our paper “MedVidDeID: Protecting Privacy in Clinical Encounter Video Recordings” was accepted to the Journal of Biomedical Informatics (JBI).
- March 17, 2025: I was accepted into the Penn AI Fellows program.
- March 4, 2025: Our paper “Towards a Real-time Clinical Agenda Setting System for Enhancing Clinical Interactions in Primary Care Visits” was presented by PhD student Sameer Bhatti at the AAAI Workshop on Large Language Models and Generative AI for Health (GenAI4Health).