With a little help from artificial intelligence, doctors could soon monitor and diagnose patients from far away.
Researchers at CU Boulder are currently testing ways to use machine learning to psychiatry. They have developed a speech-based mobile app to organise a patient's mental health status and put it into categorize.
"We are not in any way trying to replace clinicians," Peter Foltz, a research professor at the Institute of Cognitive Science and co-author of a new paper in Schizophrenia Bulletin, told a news portal. Adding, "But we do believe we can create tools that will allow them to better monitor their patients."
Millions of people across the globe live with some form of mental illness. Many of them live in areas where psychologists may not be readily available. Time and cost are also factors many people don't seek help. "Even when a patient does make it in for an occasional visit, therapists base their diagnosis and treatment plan largely on listening to a patient talk - an age-old method that can be subjective and unreliable," study co-author Brita Elvevåg, a cognitive neuroscientist at the University of Tromsø, Norway, told a news portal. Adding, "Humans are not perfect. They can get distracted and sometimes miss out on subtle speech cues and warning signs."
That is why Elvevåg and Foltz decided to work together with the aim of developing a machine learning technology that can examine changes in day-to-day speech to be able to detect mental health decline. Change in tone or pace can be a sign of depression. Memory loss can also be an indication of mental health problems.
"Language is a critical pathway to detecting patient mental states," Foltz told a news portal. Adding "Using mobile devices and AI, we are able to track patients daily and monitor these subtle changes."
The new mobile app asks patients to answer a few questions for about 10 minutes. While talking into their phone, they have to tell a short story, repeat a story they have to listen to and answer questions relating to their emotional state.
In a trial test, clinicians were asked to listen and assess speech samples of 225 participants. Their notes were compared to the results obtained from the machine learning tool. "We found that the computer's AI models can be at least as accurate as clinicians,"Foltz told a news portal.
Researchers hope the AI system they are working on can one day be used by therapists. Doctors could also be notified if the app signals a disturbing change. "Patients often need to be monitored with frequent clinical interviews by trained professionals to avoid costly emergency care and unfortunate events," Foltz told a news portal. Adding "But there are simply not enough clinicians for that."
Picture Courtesy: Google Images