Temporal Data Analytics for Clinical Decision Support
Lucia Sacchi
February 9, 2022, Wednesday, 3:00 PM - 4:00 PM EST
Abstract
The data collected during the medical history of a patient are often longitudinal in nature. Moreover, they can come from different sources (hospital, home monitoring), and be collected with different purposes (clinical or billing purposes), thus resulting in heterogeneous formats and granularities. The definition of analytics techniques able to deal with these unique characteristics of the data is important for enhancing decision support with features that take into account the temporal evolution of the disease and the involved processes of care. This talk will explore temporal analytics techniques applied to the analysis of heterogeneous data for clinical decision support. In particular, we will consider electronic temporal phenotyping as the identification of clinically meaningful event sequences from data that have been collected over time, and its application to diabetes, cancer, and COVID-19.
Bio
Dr. Lucia Sacchi is in Department of Electrical, Computer and Biomedical Engineering, University of Pavia, Italy. She obtained a Master's degree in Computer Science and Engineering, and a Ph.D. in Bioengineering and Bioinformatics, both from the University of Pavia. Her research interests include clinical decision support systems, personalized medicine, clinical data mining, big data analytics, temporal abstractions, and temporal association rule mining.