Detecting human emotions with RADIO WAVES: Scientists develop AI technology using wireless signals to reveal changes in heart rate and tell how someone is feeling
- AI technology uses wireless signals in people to tell how they are really feeling
- Researchers used neural network to understand changes physical changes
- Participants shown range of emotion-evoking videos and reactions analysed
- Hopes it could help medical professionals understand how patients are feeling
Scientists have developed AI technology using wireless signals to reveal changes in heart rate and tell how someone is feeling.
The new research, conducted at Queen Mary University of London and led by Yang Hao, uses an AI called a neural network to understand what the changes mean.
According to Defense One, it could be a breakthrough piece of research that could help figure out if situations, such as those seen in the military, are a threat.
Researchers at Queen Mary University of London used a neural network to better understand how someone is feeling by identifying ‘hidden’ changes in heart and breathing rates. Picture: Stock
Created by computer scientists, artificial neural networks are made up of various nodes — equivalent to biological neurons — that process and pass on signals.
The network can change as it is used — such as by increasing the weight given to certain nodes and connections — allowing it to ‘learn’ as it goes along.
For example, given a set of cat pictures to study, a network can learn to pick out characteristic cat features on its own — and so tell them apart from other animals.
The interconnected computational system is loosely modelled on the human brain.
For the latest research, participants were shown a range of videos which evoked different emotions, including anger, sadness and joy.
Radio waves, such as those typically transmitted by radar and Wi-Fi, were sent in the participants direction and then and antenna measured the changes in signal as it bounced back.
The universe could be a neural network — an interconnected computational system similar in structure to the human brain — a controversial theory has proposed. Picture: Stock
The changes in the waves were then analysed based on the subtle movements the participants made as they watched the different clips.
This allowed the neural network to identify patterns in the participant’s heart and breathing rate according to which emotion was being evoked.
Once the data from all the participants was collected it was plugged into the neural network to see if it could successfully identify the video being watched based on the subtle changes.
The network had a 71 per cent success rate at correctly identifying the participants’ emotion.
According to Engineering & Technology, while such research is typically used in the psychological and neuroscientific experiments, it would could also have a significant impact on managing health and wellbeing.
It is hoped the research will be able to help medical professionals get a better understanding of their patients emotions which may be ‘hidden’.