Death by suicide is the seventh of the leading death cause worldwide. The recent advancement in Artificial Intelligence (AI), specifically AI application in image and voice processing, has created a promising opportunity to revolutionize suicide risk assessment. Subsequently, we have witnessed fast-growing literature of researches that applies AI to extract audiovisual non-verbal cues for mental illness assessment. However, the majority of the recent works focus on depression, despite the evident difference between depression signs and suicidal behavior non-verbal cues. In this paper, we review the recent works that study suicide ideation and suicide behavior detection through audiovisual feature analysis, mainly suicidal voice/speech acoustic features analysis and suicidal visual cues.

Recent scientific progress has allowed us to begin decoding the significance of many different patterns of activity in the brain. Researchers have begun to understand patterns associated with disorders such as depression, in hopes of correcting it. Other research has zeroed in on how language and speech is signaled in the brain. In one often-cited experiment, researchers were even able to convert the MRI readouts of the test subjects' brains into approximate renditions of the movie clips shown to participants. Now new research seeks to use artificial intelligence to identify people suffering from suicidal thoughts based on brain scans alone. In a paper published this week in Nature Human Behavior, researchers from Carnegie Mellon University looked at 34 participants, half of whom were experiencing suicidal thoughts.

Suicide is the second-leading cause of death among young people between the ages of 15 and 34 in the United States, and clinicians have limited tools to identify those at risk. A new machine-learning technique documented in a paper published today in Nature Human Behaviour (PDF) could help identify those suffering from suicidal thoughts. Researchers looked at 34 young adults, evenly split between suicidal participants and a control group. Each subject went through a functional magnetic resonance imaging (fMRI) and were presented with three lists of 10 words. All the words were related to suicide (words like "death," "distressed," or "fatal"), positive effects ("carefree," "kindness," "innocence"), or negative effects ("boredom," "evil," "guilty").

Suicide is the second-leading cause of death among young people between the ages of 15 and 34 in the United States, and clinicians have limited tools to identify those at risk. A new machine-learning technique documented in a paper published today in Nature Human Behaviour (PDF) could help identify those suffering from suicidal thoughts. Researchers looked at 34 young adults, evenly split between suicidal participants and a control group. Each subject went through a functional magnetic resonance imaging (fMRI) and were presented with three lists of 10 words. All the words were related to suicide (words like "death," "distressed," or "fatal"), positive effects ("carefree," "kindness," "innocence") or negative effects ("boredom," "evil," "guilty").

New research published in the journal Suicide and Life-Threatening Behavior shows how machine learning can help identify suicidal behavior using a person's spoken or written words. The technology was able to pinpoint which participants in the study were suicidal, mentally ill but not suicidal, or neither in the vast majority of cases. John Pestian and a team of researchers studied 379 patients from emergency departments and inpatient and outpatient centers at three locations between Oct. 2013 and March 2015. The patients, who were classified as suicidal, mentally ill but not suicidal, or neither (serving as the control group), answered standardized behavioral rating tests and took part in a semi-structured interview in which they were asked five open-ended questions such as "Do you have hope?" and "Are you angry?" to stimulate conversation. The researchers then pulled verbal and non-verbal language (e.g., laughs, sighs, etc.) from the gathered data and used machine learning algorithms to analyze it.