Skip to main content

My research focuses on the development of statistical models and computational techniques/algorithms for complex high dimensional temporal spatial data analysis and their applications for medical research. These applications include but not limited to the identification of careless survey responses, exploring heterogeneous population with homogenous structures/relationship, development of objective biomarker and dynamic risk prediction tools. One of my most recent focus is on mobile device data analysis and its applications. Mobile device data and mobile device based behavior intervention have become increasingly popular across many different research fields, such as mental health research. The large amount of mobile sensor data collected from various studies provide a great opportunity for the development of early risk prediction tool and objective biomarkers for mental health outcomes, such as depression and PTSD. However, these mobile device data, such as heart rate and keystroke data, are often high dimensional, irregular, intensive and multi modal. Extracting informative features from these data not only requires in-depth domain knowledge but also advanced analytic models/algorithms. Our group have conducted several studies based on mobile device data and have identified a large number of objective passive behavior biomarkers for various mental health disorders, such as PTSD, depression, pain and somatic symptoms.


UNC AFFILIATIONS:

Anesthesiology

CLINICAL/RESEARCH INTERESTS:

Mental illness, Quality of Life