Capturing and Analyzing Vocal Biomarkers for Depression Research

A biomarker is a measurable characteristic that can be an indicator of health and well-being

Potential of Vocal Biomarkers in Medical Research

A biomarker is a measurable characteristic that can be an indicator of health and well-being. Various biomarkers have been used in frequently in medical research but more recently researchers have been focusing on the untapped potential of certain biomarkers such as vocal samples and facial expressions and their contribution in the study of neuropsychological disorders and depression.

Depression and Vocal Changes

Widely prevalent (estimated as affecting 300 million globally1) yet not completely understood and often left untreated due to lack of awareness, depression has been known to lead to serious health complications and, at times, the loss of precious lives.

In a path-breaking initiative to understand depression better, a team of researchers from MIT conducted the VoiceUp research study to understand the neurological basis of vocal changes due to depression.

Going Mobile for the MIT VoiceUp Study

The MIT VoiceUp objective was to capture responses to study related survey questions and to collect voice samples from eligible study participants, through a mobile app. Using a mobile app to conduct the study was imperative for two reasons – to allow the participants to take part in a comfortable and familiar environment and to accurately capture vocal samples using the mobile phone’s audio recording capability.

BTC worked with the MIT team to design and build the VoiceUp mobile app. The mobile app administered surveys and voice-capture tasks to the participants and collected data for the study. The vocal sample data collected from the mobile app was then fed into MIT’s state-of-the-art neuroimaging and vocal analysis algorithm to explore the underlying brain mechanisms that cause voice changes in adults with depression.  The study participants were also provided with objective feedback on their mental health based on the results of the vocal analytics run on their study responses.

Research meets Technology

The mobile app was built as on the Ionic platform using the Ionic Research Kit for both the iPhone and Android phones. Study enrolment, consent and withdrawal was handled and tracked for compliance purposes and the mobile app had a 2-factor authentication for additional security. The response data was stored on a secure cloud instance. The BTC Compass framework was instrumental in getting the mobile app quickly with high quality and robust security.

Whether it was the study’s vision, the domain knowledge needed or the technology used to make it happen, this project proved to truly be in the ‘BTCian space’!