Effective detection of anxiety among adolescents

In the United Kingdom, one-in-six young people experience an anxiety condition in their life, including obsessive compulsive disorder, social anxiety and shyness, exam stress, generalised worry or panic attacks. Research reveals that over half of all mental ill health starts by age 14 and 75% develops by18. Currently, there is an increase in prevalence of mental health problems at the ages of 16-19 [1].

Anxiety affects people in different ways. If prolonged, the human body is not signalled to return to normal functioning. This may weaken the immune system, leaving the sufferer vulnerable to viral infections. Continuous anxiety also sacrifices excretory and digestive systems, leading to vomiting, diarrhoea or constipation. Other symptoms include muscle tension, headaches and insomnia. Untreated anxiety can have a significant impact on school learning, confidence and personal relationships [2, 3].

To effectively prevent significant consequences, early detection of anxiety must be achieved before professional counselling can be efficiently undertaken. For this purpose, our aim is to develop novel artificial intelligence (AI) tools to identify and detect anxiety among school students through detecting physiological responses, using sensors such as Fitbit and BlueBeach to measure heart beats and skin moisture.

This is a multi-disciplinary research project, pulling together expertise in Informatics, Psychology, Sociology, and Media and Communication at Leicester. The following specific objectives form the core activities of the proposed research programme:
(1) To develop an effective questionnaire for determining if a student experiences a certain form of anxiety (e.g. exam or social).
(2) To develop a new AI framework for separating typical and atypical anxiety among students.
The expected outcome of the work is a novel automated system of identifying unhealthy anxiety among students. In regard to the dissemination of our work, in the short term, the student to be recruited will aim to publish at least three papers in top journals, such as Proceedings of the National Academy of Sciences of the United States of America. The student will present the research findings at relevant leading conferences such as International Conference on Data Mining, and publicise them via press releases and social media (including LinkedIn). In the longer term, the student will disseminate the developed tools through social and health channels.