Mobile app uses voice analysis to detect mood changes in bipolar patients

An experimental smartphone application devised by a University of Michigan research team monitors subtle changes in the voices patients diagnosed with bipolar disorder and identifies mood changes.

The changes detected could be an early warning indicator of a need for care, according to researchers. The project, called PRIORI, aims to uncover a biological marker to prioritize bipolar care to those who need it to stabilize mood swings.

Researchers say the initial project findings in studying 60 American patients are encouraging on several fronts.

"These pilot study results give us preliminary proof of the concept that we can detect mood states in regular phone calls by analyzing broad features and properties of speech, without violating the privacy of those conversations," Zahi Karam, a postdoctoral fellow and specialist in machine learning and speech analysis at U-M. "As we collect more data the model will become better, and our ultimate goal is to be able to anticipate swings, so that it may be possible to intervene early."

Patients' voice patterns are monitored by an Android app running in the background of a voice call. The app analyzes many characteristics of the sounds and silences of each conversation.

"The ability to predict mood changes with sufficient advance time to intervene would be an enormously valuable biomarker for bipolar disorder," U-M psychiatrist Melvin McInnis, who also worked on the project, said.

The research arrives at a time when some believe mHealth apps are poised for a mainstream breakthrough. A recently published Mobiquity report states 70 percent of consumers are using fitness and health monitoring apps on a daily basis and 63 percent plan to expand use over the next five years. The report also reveals 30 percent describe mHealth app use as more valuable than making or receiving calls.

Data published in January by researchers from Beth Israel Deaconess Medical Center in Boston indicates the feasibility of utilizing patient-owned smartphones to monitor mental health conditions. The rapid growth and adoption of smartphones, in particular, offers a promising mechanism for the experience sampling method (ESM) collection, according to the article's authors.

For more information:
- read the announcement

Related Articles:
Smartphone apps are feasible tools for monitoring mental health
Report: mHealth apps primed to hit mainstream this year
Mayo app offers round-the-clock medical concierge subscription service
Gamification is a winning approach for mHealth tech
What mHealth app makers need to consider before development
Smartphone proves to be a useful tool for teletrauma program
Smartphone app proves valuable for cardiac patients

Read more on