Dr. Xiaozhong Liu is a Professor of Data Science and Computer Science at Worcester Polytechnic Institute. His research focuses on artificial intelligence, data science, and their applications in healthcare, with particular interests in longitudinal patient modeling, chronic disease management, clinical decision support, human-AI collaboration, and trustworthy personalized AI.
His recent work investigates how multimodal longitudinal data, including electronic health records, wearable sensing, patient interactions, and real-world behavioral data, can be used to model evolving patient states, identify emerging risks, support treatment adherence, and enable timely clinical interventions. Current research directions include AI-assisted triage and decision-making, continuous monitoring for chronic disease, cancer treatment toxicity surveillance, digital twin models for longitudinal care, and adaptive allocation of responsibilities among AI systems, caregivers, nurses, and clinicians.
Methodologically, his work spans large language models, memory and personalization, machine learning, causal and decision modeling, information retrieval, graph learning, and privacy-aware AI. He has published more than 150 peer-reviewed papers and has led or participated in research funded by NIH, NSF, NSA, and other organizations.
Before joining WPI, Dr. Liu was a faculty member at Indiana University Bloomington and served as a Research Director and Senior Consultant at Alibaba DAMO Academy, where he led a multidisciplinary AI/NLP team and helped develop large-scale production AI systems serving more than 170 million API calls per day.
I’m actively looking for PhD students with good experience in NLP and/or deep learning.
Dr. Xiaozhong Liu is a Professor of Data Science and Computer Science at Worcester Polytechnic Institute. His research focuses on artificial intelligence, data science, and their applications in healthcare, with particular interests in longitudinal patient modeling, chronic disease management, clinical decision support, human-AI collaboration, and trustworthy personalized AI.
His recent work investigates how multimodal longitudinal data, including electronic health records, wearable sensing, patient interactions, and real-world behavioral data, can be used to model evolving patient states, identify emerging risks, support treatment adherence, and enable timely clinical interventions. Current research directions include AI-assisted triage and decision-making, continuous monitoring for chronic disease, cancer treatment toxicity surveillance, digital twin models for longitudinal care, and adaptive allocation of responsibilities among AI systems, caregivers, nurses, and clinicians.
Methodologically, his work spans large language models, memory and personalization, machine learning, causal and decision modeling, information retrieval, graph learning, and privacy-aware AI. He has published more than 150 peer-reviewed papers and has led or participated in research funded by NIH, NSF, NSA, and other organizations.
Before joining WPI, Dr. Liu was a faculty member at Indiana University Bloomington and served as a Research Director and Senior Consultant at Alibaba DAMO Academy, where he led a multidisciplinary AI/NLP team and helped develop large-scale production AI systems serving more than 170 million API calls per day.
I’m actively looking for PhD students with good experience in NLP and/or deep learning.
Scholarly Work
Adaptive Multi-Task Positive-Unlabeled Learning for Joint Prediction of Multiple Chronic Diseases Using Online Shopping Behaviors [Expert Systems with Applications 2022]
National Institutes of Health
Venture Capital
National Security Agency
National Science Foundation
National Science Foundation
Department of Defense