Signal processing & Artificial Intelligence for Time variant Health data
Welcome to the SATH lab, where we develop cutting-edge signal processing and machine learning approaches to using wearable devices, mobile phones, and clinical devices to improve health outcomes, access to care, and quality of life.
We’re Changing the Way the World Thinks About Personal Health
How can we better understand sleep, stress, & mental health, using data-driven tools?
Can passively collected lifestyle data quantitatively measure depression, stress & anxiety?
How does the timing of health-related interventions affect their efficacy?
Can we quantify the conditions and susceptibilities of the brain by measuring it's electrodynamic state?
Can we revolutionize clinical treatment plans by directly measuring the efficacy of medications in the brain?
We Develop Tools and Approaches That Enhance Our Ability to Discover New Digital Biomarkers & Phenotypes
How can we cluster multidimensional, multiscale, and multivariate, longitudinal time series data?
What approaches can we use to maintain prediction quality despite intermittently missing longitudinal data?
Which change point detection algorithms are most effective for interpreting smartphone/smartwatch/EEG data?
How does signal artifact affect the mathematical values of key nonlinear measures (e.g. entropy, RQA, etc.)?
How can we use deep learning methods to identify micro-activity in high-frequency longitudinal signals?
We are currently not hiring postdoctoral fellows, software developers or any support staff. Please stay tuned, as this may change in the future!