Oct 31, 2017

In small study, machine-learning tech can identify suicidal tendencies

A brain-scanning MRI machine. Photo: Keith Srakocic / AP

New artificial intelligence, detailed in a paper published today in Nature Human Behavior, has helped researchers to teach machines to spot suicidal tendencies and help identify those at risk, per MIT Technology Review.

Why it matters: The success of the study reveals the great potential there is for AI to aid in psychiatry, in addition to the medical developments that AI has already contributed to, such as in detecting tumors and predicting depressive disorders.

Details of the study, as detailed by MIT Technology Review:

Researchers analyzed 34 young adults, even split between suicidal participants and a control group. Each person was sent through a functional magnetic resonance imaging (fMRI) and given three lists of 10 words all relating to suicide (such as "death" or "distressed"), positive effects (such as "kindness" or "innocence"), or negative effects (such as "evil" or "guilty").
They studied five specific brain locations that were determined to be the best marks to distinguish the suicidal patients from the controls. Using those locations and words, the machine identified 15 of the 17 patients with suicidal behavior and 16 of 17 control subjects.
The suicidal participants were then split into two groups, those who had attempted suicide and those who hadn't, and the machine was trained to distinguish the differences.

Go deeper

Mike Allen

40 mins ago - Technology

Trump and Zuckerberg share phone call amid social media furor

Photo: Alex Wong/Getty Images

In the week that President Trump took on social media, Axios has learned that he had a call Friday with Facebook CEO Mark Zuckerberg that was described by both sides as productive.

Why it matters: With the White House and Twitter at war, Facebook has managed to keep diplomatic relations with the world's most powerful social-media devotee.

Go deeper (<1 min. read)

Sara Fischer

46 mins ago - Economy & Business

Twitter, Google lead chorus of brands backing George Floyd protests

Illustration: Lazaro Gamio/Axios

Twitter and Google are among the dozens of brands over the past 24 hours that have taken public stances in favor of Americans protesting racial equality. Some companies have changed their logos in solidarity with the movement, while others have pledged money in support of efforts to address social injustice.

Why it matters: The pressure that companies feel to speak out on issues has increased during the Trump era, as businesses have sought to fill a trust void left by the government. Now, some of the biggest companies are quickly taking a public stand on the protests, pressuring all other brands to do the same.

Go deeper (2 min. read)

Jacob Knutson

2 hours ago - Politics & Policy

NYPD commissioner: "I'm extremely proud" of officers' response to protests

New York City Police Commissioner Dermot Shea in February. Photo: Yana Paskova/Getty Images

New York City Police Commissioner Dermot Shea said in a public statement Sunday that he is "extremely proud" of the New York City Police Department's response to protests over the death of George Floyd Saturday night, writing: "What we saw in New York City last night and the night before was not about peaceful protest of any kind."

Why it matters: New York City residents captured several instances of police officers using excessive force against demonstrators. In one video, two NYPD SUVs are seen ramming into protesters who were blocking a road and throwing traffic cones at the vehicles.

Go deeper (1 min. read)