As a direct PhD student in the Electrical Engineering faculty at the Technion–Israel Institute of Technology, Tal Shnitzer dealt with the problem of how to extract the essence of data when facing highly complex signals, e.g. medical data. She developed data-driven methods for time-series filtering and for sensor fusion based on manifold learning techniques, which reveal the underlying geometrical structure of the data. The power and capabilities of these methods were demonstrated in several medical applications, including fetal heart-rate recovery from non-invasive maternal ECG recordings and Alzheimer’s disease identification from resting state EEG recordings.
As a postdoc, Dr. Shnitzer continues her interest in geometric problems and medical applications as part of the Geometric Data Processing Group at the MIT Computer Science and Artificial Intelligence Laboratory. She plans to research problems in computer graphics and machine learning, aimed at solving challenges in complex medical data analysis.
An award-winning teacher, Dr. Shnitzer consistently receives outstanding scores and rave reviews from her students.