Anyone who has ever been affected by suicide knows that, even to close friends and loved ones, the signs can be difficult to spot — with often tragic results. To try and help with this very serious problem, researchers from Carnegie Mellon University and the University of Pittsburgh have been awarded a $3.8 million grant from the National Institute of Mental Health to apply some cutting-edge technology. They hope to use brain imaging and machine learning technology to aid clinicians in detecting and treating suicidal patients.
“We examine brain activation patterns that occur while people are thinking about suicide-related and emotional concepts,” David Brent, a professor in the department of psychiatry at the University of Pittsburgh, told Digital Trends. “We then use machine learning to identify the words and activation patterns associated with those words that best discriminate between suicidal and non-suicidal individuals.”
The ability to accurately scan someone to quickly determine whether they might pose a suicide risk may still be a way off, but the researchers’ previous work does at least suggest this is a possibility. In a study carried out last year, they demonstrated that it is possible to use this technology to tell whether a person is considering suicide based on the way they think about death-related topics. The new NIMH grant will help take this work forward.
“The first study was relatively small, had no non-suicidal psychiatric participants, and was cross-sectional,” Brent continued. “This study will be larger, [and] will repeat the procedure to see if fluctuations in neural signatures are related to fluctuations in suicidal ideation, and see how we can predict future suicidal behavior. In addition, we will be developing and testing a peripheral measure that we hope will correlate with neural findings, but can be used in a clinician’s office without needing to do fMRI.”
Ultimately, the researchers are hoping to establish a more objective way to assess suicide risk, rather than having to rely on self-reporting. They also believe that this work will be able to reveal more about the ways in which suicidal people think about death, which could be used to more precisely guide psychotherapy.
- Freaky new A.I. scans your brain, then generates faces you’ll find attractive
- The best therapy apps for Android and iOS
- Image-recognition A.I. has a big weakness. This could be the solution
- Exoskeletons with autopilot: A peek at the near future of wearable robotics
- Why teaching robots to play hide-and-seek could be the key to next-gen A.I.