Vasily Giannakeas, PHD, MPH

Combining his expertise in engineering, epidemiology and public health, Dr. Giannakeas is positioned to explore how digital health solutions will change the landscape of epidemiological research and address emerging issues in public health. There is a shift in the focus from individual primary data collection to the use of large databases collected for administrative purposes. This transformation brings a unique set of challenges, as many data sets are intended for reasons other than research. Dr. Giannakeas will address these challenges by developing innovative solutions for using digital health data in epidemiologic research, including creating novel approaches for solving data-related problems, and identifying and addressing biases in observational studies.

Dr. Giannakeas is involved in the development of a Learning Health System at Women’s College Hospital. Many questions related to patient care go unanswered because of the time and costs required to run clinical trials. A Learning Health System allows for the evaluation of routinely collected clinical data within a research framework to improve patient outcomes and patient satisfaction. Dr. Giannakeas collaborates with physicians in many of the medical departments at Women’s College Hospital to identify and address key research questions in clinical practice.

Much of his previous work focuses on breast and ovarian cancer. Dr. Giannakeas has leveraged large administrative health databases to address novel questions related to cancer. His research interests range from early cancer detection to cancer survivorship. He recently developed a cancer prediction model that uses historical laboratory data and medical records to identify profiles of individuals at a high risk of cancer up to 5 years before disease onset. Dr. Giannakeas aims to evaluate the clinical utility of this prediction tool by identifying high-risk patients in a family practice setting and offering additional cancer screening where appropriate, with the goal of early detection and improved survival.