The research team at Dartmouth University announces the development of the first mobile application utilizing artificial intelligence in conjunction with facial image processing software to reliably detect the onset of depression before the user even becomes aware of it.
Illustration Image Pixabay
Named MoodCapture, this app uses the phone's front camera to capture facial expressions and the surroundings of an individual during their regular use, then evaluates the images looking for clinical signs associated with depression. In a study involving 177 individuals diagnosed with major depressive disorder, the app correctly identified early symptoms of depression with an accuracy of 75%.
These findings suggest that the technology could be made publicly available within the next five years with further developments. The researchers, based at the Department of Computer Science and the Geisel School of Medicine, have published their findings on the pre-publication database
arXiv.
"This is the first time that natural 'in-the-wild' images have been used to predict depression," states Andrew Campbell, a professor of computer science and author of the study. MoodCapture operates by capturing images of the participants over 90 days without them knowing when this occurs. The photos are analyzed to detect specific facial expressions and environmental features associated with depression. The app makes connections between the expressions and important background details to predict the severity of depression.
The research, funded by a grant from the National Institutes of Mental Health, explores the use of deep learning and passive data collection to detect symptoms of depression in real-time. It also builds on a previous study from 2012 conducted by Campbell's laboratory on passive and automatic data collection from the participants' phones. The MoodCapture app offers a promising glimpse into the future of technology in supporting mental health, providing ongoing, non-intrusive mood assessment, with the potential for early detection and treatment of depression symptoms.
Article author: Cédric DEPOND