Document Type

Thesis

Publication Date

12-2019

Keywords

machine learning, deep neural network, virtual reality exposure-based treatment, emotion recognition, physiological signals, multimodal

Abstract

Post-Traumatic Stress Disorder is a mental health condition that affects a growing number of people. A variety of PTSD treatment methods exist, however current research indicates that virtual reality exposure-based treatment has become more prominent in its use.Yet the treatment method can be costly and time consuming for clinicians and ultimately for the healthcare system. PTSD can be delivered in a more sustainable way using virtual reality. This is accomplished by using machine learning to autonomously adapt virtual reality scene changes. The use of machine learning will also support a more efficient way of inserting positive stimuli in virtual reality scenes. Machine learning has been used in medical areas such as rare diseases, oncology, medical data classification and psychiatry. This research used a public dataset that contained physiological recordings and emotional responses. The dataset was used to train a deep neural network, and a convolutional neural network to predict an individual’s valence, arousal and dominance. The results presented indicate that the deep neural network had the highest overall mean bounded regression accuracy and the lowest computational time.

Comments

A Thesis presented to The Faculty of Applied Science and Technology, School of Applied Computing in partial fulfillment of requirements for the degree of Bachelor of Applied Computer Science Mobile Computing.

Faculty

Faculty of Applied Science & Technology (FAST)

Terms of Use

Terms of Use for Works posted in SOURCE.

Creative Commons License

Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Original Publication Citation

Dass, N. (2019). Exploring emotion recognition for VR-EBT using deep learning on a multimodal physiological Framework (Unpublished thesis). Sheridan College, Ontario, Canada.

Share

COinS