Numerous innovations and product enhancements have come out of science laboratories over the past few years. One of these, wireless emotion-detection technology, holds much promise in diagnosing complex conditions like depression in the near future.
The Massachusetts Institute of Technology’s Computer Science and Artificial Intelligence Laboratory has rolled out a wireless device called EQ-Radio. The innovation can tell if a person is sad, excited, happy, or angry, by measuring breathing and heart rhythms.
Dina Katabi, director of the MIT Wireless Center, had shared credits for the innovation with MIT Assistant Professor Fadel Adib (recognized as MIT Technology Review’s world’s top 35 innovators under 35 and who also landed among Forbes’ list of 30 under 30) and Ph.D. student Mingmin Zhao.
USE OF MACHINE LEARNING CONCEPTS
The device can analyze small variations or subtle changes in breathing and heartbeat intervals. The research team used concepts from machine learning. Participants were made to recall memories that produced a range of emotions and were asked to secretly elicit an emotion, serving as part of the test phase.
The test included 12 participants and 40 tests. Katabi and her team members wrote a paper on the topic detailing the measurements used to determine emotion. When the signals showed low arousal and negative affect, the device registered the emotion as sad (high arousal and positive affect were interpreted as excited).
By recovering measurements of the heart valves opening and closing at a millisecond time-scale, the device can literally detect if someone’s heart skips a beat, Adib said. The paper was presented at the recent MobiCom 2016, the Annual International Conference on Mobile Computing and Networking held in New York City.
The implications of the innovation extend beyond emotion recognition. Real-world applications of the EQ-Radio can be centered on improving the quality of life of people who may suffer from extreme emotions, depression and other mental illnesses as well as behavioral disorders.