Depression is a reality for many people, but due to the wide scope of symptoms and varying degrees of severity, it is often difficult to diagnose.
Researchers at the Computer Science and Artificial Intelligence Laboratory are attempting to combat this difficulty. They aimed at making diagnosis simpler for doctors and individuals by creating an algorithm that when taught to a machine, can detect depression simply by listening to the tone of your voice.
The technology is in its infancy, so while it promises to be a wonderful tool, there are still some hurdles to overcome. The algorithm is based on a study that used text and audio data from 142 interviews with patients; 30 of the patients had been diagnosed with depression, and the machine was able to correctly identify the other patients with a 77 percent success rate.
Tuka Alhanai, a researcher from the laboratory and one of the developers of the algorithm, said: "If you want to deploy models in a scalable way, you want to minimize the amount of constraints you have on the data you’re using."
