Treating Depression through Mobile Technology

A team of researchers from UConn and UConn Health have received a $1M grant from the National Institute of Mental Health and Department of Health and Human Services to develop a system using mobile health technologies and machine learning to provide clinicians with better assessments of depression symptoms.

Young woman looks at iphone

Image by rawpixel from Pixabay.

One size does not fit all. This is true when it comes to clothing, but more importantly, healthcare.

We all have different bodies that react differently to the same treatment, meaning doctors often need to try multiple pathways for treatment before finding something that works for a given patient. Even then, something could stop working over time, further proving why personalization and up-to-date monitoring in healthcare is essential. This is especially true when it comes to treating depression.

University of Connecticut professor of computer science and engineering Bing Wang is the lead principal investigator on a new $1 million grant from the National Institute of Mental Health and Department of Health and Human Services to develop a system that can help provide personalized treatment data for people suffering from depression and that is informed by innovative technology.

Jinbo Bi, associate head of the Department of Computer Science and Engineering, Alexander Russell, professor of computer science and mathematics, and Jayesh Kamath, associate professor of psychiatry at UConn Health, are co-PIs on this project.

The system the researchers will develop, DepWatch, will use mobile health technologies and machine learning to provide clinicians with objective, timely, and accurate assessments of depression symptoms. These assessments will help inform clinicians’ decisions by providing them with more data about how patients are reacting to their current treatment regimes.

DepWatch collects data from smartphones and wristbands without any user interaction. The system will also allow users to input ecological momentary assessments (EMAs). EMAs involve users entering data about their symptoms and experiences as they happen. This eliminates possible problems with clinicians having to rely on patients’ ability to remember how they were feeling at a particular moment weeks later at a checkup appointment. DepWatch can also track medication adherence and safety information.

All this information will be fed into machine learning models that will develop a weekly assessment of patient symptoms and predict their responses to treatment over time. Clinicians can then use this data to make better-informed decisions about what kind of treatment will work best for their patients.

The project will first gather data from 250 adult participants with depression to build machine learning models in DepWatch. In the second phase, the project will recruit 128 adult patients treated by a group of clinicians, with the goal of testing the feasibility and effectiveness of the DepWatch system.

“This project will take a significant step forward in providing efficient, patient-centric care in treating depression,” Wang says. “It will help advance personalized depression treatment by identifying early patients who are failing treatments and assist providers to take necessary actions before patients drop out of treatment altogether.”

Wang leads the Computer Networking Research Group at UConn. Her research group works broadly in computer networking, including ubiquitous computing, multimedia streaming, network management, performance modeling and evaluation, and network security. Wang received her Ph.D. in computer science from the University of Massachusetts, Amherst.

This grant is NIH No.: 1R01MH119678-01

Follow UConn Research on Twitter & LinkedIn.