Nonprofit Uses Machine Learning to Analyze Veteran Device Data and Prevent Military Suicides
The Black Box Project, an initiative by Stop Soldier Suicide, a nonprofit organization, is utilizing machine learning technology to analyze data from the devices of veterans who died by suicide. By collecting cell phones, tablets, and laptops left behind by these veterans, the organization aims to gain insight that could help prevent future military suicides. The project is named after the black boxes found on planes, which assist in determining the cause of aviation incidents.
According to the Department of Veterans Affairs, veterans are 57% more at risk for suicide than non-military adults, and around 6,000 veterans die by suicide each year. Stop Soldier Suicide CEO Chris Ford believes that by gathering unique data from those who passed away by suicide, they may be able to recreate the last year of their lives and potentially prevent needless suffering and loss of life.
David McDaniel, whose son Conor died by suicide at the age of 26, has experienced the devastating impact of military suicides firsthand. Conor joined the Army after high school and appeared to be in good spirits during his deployment to Afghanistan. However, upon returning home, David noticed a shift in Conor’s behavior towards anger, frustration, and gloominess. Conor’s tragic death prompted David to support the Black Box Project, as he believes that identifying patterns in online searches could help prevent similar tragedies. For example, if a veteran searches for information on suicide, an ad for Stop Soldier Suicide could potentially be triggered, providing resources and support.
The nonprofit employs machine learning, a subfield of artificial intelligence, to analyze texts, notes, and phone usage patterns of deceased veterans. By monitoring their previous sleep cycles and past geolocations, the organization aims to uncover signals that could indicate anger, desperation, or social isolation. Machine learning algorithms can process data at a faster rate compared to humans and can detect patterns that may be missed by traditional methods.
Since its launch in April 2020, the Black Box Project has received over 100 devices. Preliminary findings suggest three patterns that often precede military suicides: increased anger, changes in sleep patterns, and social isolation. Stop Soldier Suicide’s team, which includes clinical psychologists and suicide researchers, works closely with a scientific advisory committee to determine where and what to look for on the devices.
Privacy concerns regarding data collection are taken seriously by the organization. Stop Soldier Suicide does not sell or distribute the data to third parties and only uses it for research purposes to better understand risk factors. After analysis, the devices are returned intact to the families of the deceased veterans.
Despite the sensitive nature of the project, many next of kin, like Petra Oxford-Jackson, who lost her veteran husband Anthony to suicide, have willingly contributed their loved ones’ devices to the Black Box Project. Petra believes that if it could save even one life, it would be worth it.
Earlier this year, the Black Box Project secured $3 million in funding from the Department of Veterans Affairs to support its research and initiatives. The ultimate goal is to use the data gathered to determine when and where mental health resources should be targeted, such as adding more hotline staff during certain hours and ensuring calls are answered overnight.
By leveraging machine learning technology and analyzing veteran device data, Stop Soldier Suicide hopes to make significant strides in preventing military suicides. The organization’s commitment to privacy and ethical data usage ensures that the data collected serves the purpose of understanding and mitigating risk factors. With further research and development, the insights gained from the Black Box Project could potentially save lives and provide invaluable support to those who have served their country.
