Real-Time Classification of Anxiety in Virtual Reality Therapy Using Biosensors and a Convolutional Neural Network.

Virtual Reality Exposure Therapy is a method of cognitive behavioural therapy that aids in the treatment of anxiety disorders by making therapy practical and cost-efficient. It also allows for the seamless tailoring of the therapy by using objective, continuous feedback. This feedback can be obtained using biosensors to collect physiological information such as heart rate, electrodermal activity and frontal brain activity. As part of developing our objective feedback framework, we developed a Virtual Reality adaptation of the well-established emotional Stroop Colour-Word Task. We used this adaptation to differentiate three distinct levels of anxiety: no anxiety, mild anxiety and severe anxiety. We tested our environment on twenty-nine participants between the ages of eighteen and sixty-five. After analysing and validating this environment, we used it to create a dataset for further machine-learning classification of the assigned anxiety levels. To apply this information in real-time, all of our information was processed within Virtual Reality. Our Convolutional Neural Network was able to differentiate the anxiety levels with a 75% accuracy using leave-one-out cross-validation. This shows that our system can accurately differentiate between different anxiety levels.

Anxiety classification in virtual reality using biosensors: A mini scoping review.

BACKGROUND: Anxiety prediction can be used for enhancing Virtual Reality applications. We aimed to assess the evidence on whether anxiety can be accurately classified in Virtual Reality. METHODS: We conducted a scoping review using Scopus, Web of Science, IEEE Xplore, and ACM Digital Library as data sources. Our search included studies from 2010 to 2022. Our inclusion criteria were peer-reviewed studies which take place in a Virtual Reality environment and assess the user's anxiety using machine learning classification models and biosensors. RESULTS: 1749 records were identified and out of these, 11 (n = 237) studies were selected. Studies had varying numbers of outputs, from two outputs to eleven. Accuracy of anxiety classification for two-output models ranged from 75% to 96.4%; accuracy for three-output models ranged from 67.5% to 96.3%; accuracy for four-output models ranged from 38.8% to 86.3%. The most commonly used measures were electrodermal activity and heart rate. CONCLUSION: Results show that it is possible to create high-accuracy models to determine anxiety in real time. However, it should be noted that there is a lack of standardisation when it comes to defining ground truth for anxiety, making these results difficult to interpret. Additionally, many of these studies included small samples consisting of mostly students, which may bias the results. Future studies should be very careful in defining anxiety and aim for a more inclusive and larger sample. It is also important to research the application of the classification by conducting longitudinal studies.