advertisement
Home » Blogs » What is TMS? » Whoa – Can a Machine Learn to Spot Depression?

Whoa – Can a Machine Learn to Spot Depression?

You Won’t Believe This!

Meet Ellie.

Meet Ellie, the machine.Ellie is a computer program that has been developed and designed to spot depression by analysing the speech and facial expressions of a person who sits down with her – face-to-face. Ellie the computer asks questions that may reveal the Answerer’s trigger signs of depression. The answers themselves are irrelevant. In much of language the words are nowhere near as important as the way they are delivered or the body language that accompanies them – in this case the facial expressions are the key to diagnosing depression.

. Ellie the computer asks questions that may reveal the Answerer’s trigger signs of depression.

Ellie the Machine

I first heard about Ellie while driving and listening to NPR in May, 2013. Ellie was just a glimmer in her creator’s eye back then. Nonetheless, I was intrigued enough to make a mental note. She’s been in the news again, and is she interesting now. Ellie has blossomed!

“Her hair is brown and tied back into a professional-looking ponytail. She wears a blue shirt, tan sweater and delicate gold chain.” A typical conversation between Ellie and a new ‘client’ goes like this:

“So how are you doing today?” Ellie asks cautiously, trying to build rapport.

“I’m doing well,” he answers. His eyes blink.

“That’s good,” she continues. “Where are you from originally?”

“I’m from L.A.,” he tells her, and this makes her smile slightly.

“Oh!” she says with surprise in her voice. “I’m from L.A. myself!”

 Ellie is the courageous brainchild of a psychologist, Albert "Skip" Rizzo

(Ellie, on the right)

Skip Rizzo – Inventor, therapist, psychologist

Ellie is the courageous brainchild of a psychologist, Albert “Skip” Rizzo, who studied soldiers returning from the battlefield with Post Traumatic Stress Disorder. Rizzo saw a problem and wanted to provide a solution that really helped. He has compiled the reactions of soldiers as a kind of database for emotions that Ellie can draw upon to help recognize those who may be facing depression. The subtle differences in the stress levels of speech, the direction in which a subject looks when speaking or how long they smile for can give off signals related to depression.

“Everything has been thought of,” says Rizzo’s associate computer scientist Louis-Philippe Morency. For example, when patients talk, Ellie encourages them to continue talking with a well-timed “uh-huh,” just as real people do.

Depression

Depression has long been a difficult thing for doctors to diagnose with confidence. It is an even more difficult thing to predict because the signs are complex to spot and the fact is that most doctors don’t have the time to spend with their patients to notice the markers. When you see information about depression in the press or on awareness campaigns, the caution line is, “See your doctor.” But general practitioners may not be geared up to spot the signs and to provide mental health care. If your general practitioner is not able to help, then where can you turn?

Can a computer see our emotions?

In the movies, computers and robots are portrayed at being really good at processing data and similar tasks but poor at dealing with human emotions. Every person is different and they react to a variety of situations in a variety of ways. But Albert “Skip” Rizzo saw that he could develop Ellie to use data processing to analyze human emotions. If you take out the human emotion in the diagnostic stage of the process then you may get a more accurate result. Ellie doesn’t feel sorry for the subject or want to help them in any way. Ellie is simply analyzing every move, every speech pattern to predict a likelihood of depression.

The fact is that Ellie has practical applications to a point but once the potential for depression is recognized then the process is handed over to a human to provide help. Ellie is currently being used to help soldiers that have faced distressing situations with the armed forces and the work that is carried out here could prove invaluable for the mental health care of our armed forces. Skip Rizzo genuinely believes these technologies will eventually change the field of mental health.

Hello, Ellie. How are you?

Whoa – Can a Machine Learn to Spot Depression?


Dr. Scott West

About ThriveLogic TMS + NeuroHealth: In April of 2010, Dr. West brought the technology of NeuroStar TMS to Nashville, becoming the first physician in Tennessee to offer the option of Transcranial Magnetic Stimulation (TMS) for patients whose severe depression has not responded to a course of antidepressant medication or treatment for depression. ThriveLogic TMS + NeuroHealth (formerly Nashville TMS) Team offers the most experience in the Tennessee-area. We have treated 300+ patients across the U.S. and administered 12,000+ TMS treatments, plus we maintain some of the highest percentages of positive patient responses and remission rates in the industry. Hear what ThriveLogic patients have to say about their depression treatment experiences and outcomes! Explore Patient Testimonials. For more information on this and other topics related to the treatment of depression and mental health issues, contact us at (615) 712-6251 or info@thrivelogic.com or visit our website ThriveLogic.com.


No comments yet... View Comments / Leave a Comment

 

 

APA Reference
West, W. (2016). Whoa – Can a Machine Learn to Spot Depression?. Psych Central. Retrieved on June 19, 2019, from https://blogs.psychcentral.com/tms/2016/03/whoa-can-a-machine-learn-to-spot-depression/

 

Last updated: 10 Mar 2016
Statement of review: Psych Central does not review the content that appears in our blog network (blogs.psychcentral.com) prior to publication. All opinions expressed herein are exclusively those of the author alone, and do not reflect the views of the editorial staff or management of Psych Central. Published on PsychCentral.com. All rights reserved.