Image
Building Smarter Tools for Imperfect Health Data
Canyu Lei
Education
Ph.D. in Data Science, University of Virginia
M.E. in Systems Engineering 2025, University of Virginia
B.S. in Computer Science 2023, University of Science and Technology of China
Hometown
Xi’an, Shaanxi, China
For Canyu Lei, data science became compelling when she saw its potential to make incomplete real-world information useful. An undergraduate research project using sparse blood glucose measurements to infer meaningful clinical metrics showed her how rigorous analysis could support better decisions in healthcare.
Now pursuing a Ph.D. in Data Science at the University of Virginia, Lei is interested in developing AI methods that are accurate, interpretable, and grounded in scientific knowledge. Her work will explore representation learning, generative modeling, and physics-informed machine learning to better understand complex physiological systems — and ultimately help make health technologies more personalized, accessible, and clinically useful.
Q: When and how did you become interested in data science?
My interest in data science began during my undergraduate studies, when coursework and projects in artificial intelligence introduced me to data-driven problem-solving. It grew through a research project in which I used sparse and irregular blood glucose measurements to infer clinically meaningful glycemic metrics.
Seeing how thoughtful analysis could transform incomplete real-world data into actionable insights showed me that data science is not only about building models, but also about addressing meaningful societal challenges.
Q: Choosing a doctoral program is a big decision. Why did you choose UVA's Ph.D. in Data Science?
I chose UVA’s Ph.D. in Data Science because of its collaborative and interdisciplinary research environment. UVA offers the intellectual breadth and strong faculty mentorship I was seeking to expand my technical foundation while pursuing impactful research at the intersection of artificial intelligence and healthcare.
Q: What areas of research interest you and why?
I am interested in artificial intelligence for healthcare. I hope to explore representation learning, generative modeling, and physics-informed machine learning to capture complex physiological dynamics across multiple temporal and spatial scales. These areas interest me because incorporating scientific and physiological knowledge into data-driven models can produce solutions that are not only accurate, but also robust, interpretable, and clinically meaningful.
Q: What do you hope to do with your data science degree?
I hope to develop reliable data science methods that improve how complex health data are interpreted and used in practice. My long-term goal is to bridge methodological innovation and real clinical needs by creating tools that help researchers and healthcare professionals extract useful information from imperfect data. Through this work, I hope to contribute to more accessible and personalized smart health technologies and, ultimately, to better patient care.
Q: What advice would you give to prospective students considering a Ph.D. in Data Science? What do you wish you had known before starting?
I would encourage prospective students to view data science as more than learning algorithms or improving model performance. Meaningful research also requires understanding the application domain, communicating with collaborators, and remaining open to feedback.
I wish I had recognized earlier how iterative and collaborative the research process is. Unexpected results are not simply failures; they often reveal important assumptions, inspire better questions, and lead to stronger solutions.
Q: What are your initial impressions of the school, faculty, and other students? What are you most excited about, and what challenges do you anticipate?
My initial impression is that the School of Data Science is a welcoming and highly collaborative community. I have been especially impressed by the range of expertise represented by the faculty and students and their openness to interdisciplinary work.
I am most excited about learning from people with different perspectives and exploring new research directions. I anticipate that balancing methodological innovation with domain-specific needs will be challenging, but I also see that challenge as one of the most rewarding aspects of data science.
Q: What is a fun fact about yourself?
I’m a huge hamster lover.
Learn more about the Ph.D. in Data Science at the University of Virginia. Request more information, connect with Admissions, or start your application today.
