By Jason Matthews

A growing number of us now use smart devices and apps to monitor our physical health. Smartphones, smartwatches, fitness bands, and smart scales are just a few of these that help your average fitness enthusiast keep tabs on their progress and help those with health issues monitor their conditions.

Mental health has been a trending issue in the last year. But although seen by many as equally, if not more important than physical health, it has been restricted to apps that log mood and give advice to those who need it. That may be about to change though. Researchers at Queen Mary University London are investigating the possibility of using AI and wireless signals to monitor, understand, and record a person's deep emotional state.

Image Credit: cono0430 via Shutterstock / HDR tune by Universal-Sci

AI Emotion detection

Emotion detection, until now, has mainly been used by psychiatric professionals to research psychological or neurological disorders in patients. The methods used to monitor breathing, cardio electrical activity, muscular movements, and expressions typically rely on sensors being placed on the subject in various locations as well as facial recognition. The researchers involved in the study have been investigating the use of radio waves to measure the same signals remotely and process that information using 'deep learning' AI software.

Ahsan Noor Khan, a PhD student at Queen Mary and first author of the study, said: "Being able to detect emotions using wireless systems is a topic of increasing interest for researchers as it offers an alternative to bulky sensors and could be directly applicable in future 'smart' home and building environments. In this study, we've built on existing work using radio waves to detect emotions and show that the use of deep learning techniques can improve the accuracy of our results."

Previous remote emotion detection systems have been developed in the past, but they used computer algorithms and needed baseline data on the subject being monitored. Change the subject and you need to input the new baselines for that person to maintain accuracy. This subject-dependant system would be like learning the emotional signals of people in your household, but as soon as you meet someone new, you cannot tell if they are happy, sad, or angry. The deep learning approach that is being developed can assess the input data more humanly. It can make connections between various input data in a similar way to the human brain, allowing it to apply previous information to a new subject and learn what different signals mean from different people.

Achintha Avin Ihalage, a PhD student at Queen Mary, explains, "With deep learning, we've shown we can accurately measure emotions in a subject-independent way, where we can look at a whole collection of signals from different individuals and learn from this data and use it to predict the emotion of people outside of our training database."

Image Credit: vs148 via Shutterstock / HDR tune by Universal-Sci

Future technology

This technology is initially intended for use in studying deep emotions for scientific reasons and mental health treatment. But it could find its way into your home or, even one day, your smartphone.

Ahsan Noor Khan added, "We're now looking to investigate how we could use low-cost existing systems, such as Wi-Fi routers, to detect emotions of a large number of people gathered, for instance, in an office or work environment. This type of approach would enable us to classify emotions of people on an individual basis while performing routine activities. Moreover, we aim to improve the accuracy of emotion detection in a work environment using advanced deep learning techniques."

Professor Yang Hao, the project lead, continued, "This research opens up many opportunities for practical applications, especially in areas such as human/robot interaction and healthcare and emotional wellbeing, which has become increasingly important during the current Covid-19 pandemic."

Robot nurses that can report mental health concerns about a patient to doctors and smart homes that can automatically play soothing music, have a cup of tea ready, and warn your spouse when you are coming home after a bad day at the office are a long way off. However, this study may bring us one step closer to being able to monitor our mental health as effectively as we do our physical health.

Further reading: