Can social anxiety in young people be detected using physiological data from wearables?
Coding - Machine Learning - Experiment Design - User Centric Design - UX/UI - Prototyping
Social anxiety greatly affects young people’s lives, but the current solutions in place are inadequate for the rising prevalence of social anxiety. Detection of social anxiety using wearables may provide a novel way of recognising it, which reveals new opportunities for monitoring and treatment; this would be greatly beneficial for sufferers, society and healthcare services. This project is one of the first to investigate whether social anxiety in young people can be detected using physiological data collected from a wearable. The results were promising, they indicated that it could be possible to detect social anxiety using this approach.
The project received a Design Engineering Selected Innovation REcognition for social impact.
Speculative ideas for applications of detection social anxiety, ideas were generated with the help of sufferers during virtual focus groups
Social anxiety is defined as a fear of social situations where the individual might be exposed to possible scrutiny, it typically causes physiological changes. Unfortunately there is a high prevalence of social anxiety in young people at a subclinical level yet many do not receive treatment; due to lack of treatment available and lack of recognition by healthcare professionals. Some may not even seek treatment due to a fear of being negatively evaluated. If left untreated one can develop social anxiety disorder, which is one of the most common anxiety disorders; as well as comorbidities such as depression which cause even further life impairments and require more costly treatment.
Physiological data was recorded using an E4 Empatica, while thirteen young people with social anxiety participated in impromptu speech tasks. Following a supervised machine learning approach, various classification algorithms were then used to develop models for three different contexts, investigating the detection of social anxiety and it’s nature. Focus groups and interviews were also conducted with socially anxious young people to identify potential use cases for detection in this manner and to further evaluate the impact of the study and research in this area.
A participant's physiological data samples during the three experiment stages
The findings were promising, all classification models in all machine learning experiments (detection of presence of social anxiety, its nature and severity) yielded above 90% accuracy. Additionally, the focus groups further emphasised the positive impact of the study and further research in this area.
This research could transform the current approaches to diagnosing, treating and monitoring social anxiety which could have great social and economic impact. Feel free to get in contact if you would like to know more.
DISCOVERED AND VALIDATED THAT A
Predictive Physiological Indicators