| Titre : | Psychotherapy and Rehabilitation of Phobia using EEG Feedback in Virtual Systems |
| Auteurs : | Dounai Chebana, Auteur ; Abdelhakim Nahili, Auteur ; Ahmed Tibermacine, Directeur de thèse |
| Type de document : | Monographie imprimée |
| Editeur : | Biskra [Algérie] : Faculté des Sciences Exactes et des Sciences de la Nature et de la Vie, Université Mohamed Khider, 2023 |
| Format : | 1vol.(82p.) / ill.couv.ill.encoul / 30cm |
| Langues: | Anglais |
| Mots-clés: | Phobias, Virtual Reality (VR), Exposure Therapy, Virtual Reality Exposure Therapy (VRET), Machine Learning (ML), Deep Learning (DL). |
| Résumé : |
Anxiety disorders are a widely common mental disorder that affects millions of people worldwide. Phobias are the most common type. 15-20% of people worldwide experience at least one specific phobia in their lives. Acrophobia, the term given to the debilitating fear of heights, is the most prevalent specific phobia that reduces people’s quality of life due to their avoidance of feared situations. Traditional therapy usually uses Cognitive Behavioral Therapy (CBT) and medications to treat different kinds of phobias. These methods take a long time, are expensive, and sometimes fail to solve deeper psychological problems. Virtual Reality has been used in the last few decades as a new technology
for exposure therapy. In this thesis, we have implemented a VRET system based on a real-time automatic prediction of the following exposure scenario of height according to the estimated acrophobia level of subjects derived from the EEG signals. Various machine and Deep learning classifiers were used for acrophobia-level recognition based on the preprocessed EEG signals recorded from acrophobic subjects while exposing them to different heights in the VR environment. The novelty automatically adapts the exposure scenarios according to the subject’s acrophobia level. The results showed very high accuracies on both training and testing (96% for the training and 99% for the testing of the CNN model, 96% for the training and 97% for the testing of the MLP model). A Conditional Generative Adversarial Network (cGAN) was used to determine the next exposure scenario, and it showed promising results. A Kullback-Leibler Divergence of 0.0036 and a histogram intersection of 0.9661 demonstrate that the cGAN model successfully generated synthetic EEG data that resembles actual distributions. The Kolmogorov-Smirnov test with a p-value of 0.963 further established the cGAN’s proficiency in EEG data synthesis by validating the high similarity between the real and generated data. |
| Sommaire : |
Abstract i
R´esum´e ii Acknowledgements iii List of Figures vii List of Tables ix Introduction 1 1 Virtual Reality Exposure Therapy 3 1.1 Introduction . . . . . . . . . . . . . 3 1.2 Anxiety Disorders . . . . . . . . . . . . . . . .. . . 3 1.3 Phobias . . . . . . . . . . . . . . . . . . . . . .. . . 4 1.4 Specific Phobias . . . . . . . . . . . . . . . . . . . . . 4 1.5 Posttraumatic Stress Disorder . . . . . . . . . . 5 1.6 Psychiatric Disorder Therapy . . . . . . . . .. . . 6 1.7 Exposure Therapy . . . . . . . . . . . . . . . . . 7 1.7.1 In vivo Exposure . . . . . . . . . . . . . . 7 1.7.2 Imaginal Exposure (In vitro) . . . . . . 7 1.7.3 Systematic Desensitization . . . . . . . . . . . 7 1.7.4 Flooding . . . . . . . . . . . . . . . . . . . . . . 7 1.7.5 Virtual Reality Exposure . . . . . . . . . . 8 1.8 Human-Computer Interaction . . . . . . . . . 8 1.9 Virtual Reality . . . . . . . . . . . . . . . . . . . . .. . 8 1.9.1 Difference Between Virtual Reality and Traditional Media . . . . . 9 1.9.2 Expounding of Fundamental Definitions and Terminology in Virtual Reality . . . 10 1.9.3 Immersion in Virtual Reality . . . .. . 11 1.9.4 Characteristic of Immersive Virtual Reality . . . . . 13 iv1.10 Virtual Reality Exposure Therapy . . . . .. . . 13 1.11 Related Works . . . . . . . . . . . 14 1.11.1 Electrophysiological correlates of in vivo and virtual reality exposure therapy in spiderphobia . 14 1.11.2 Automatic adaptation of exposure intensity in VR acrophobia therapy, based on deep neural networks . . . . . . . . . . . . . . . . 15 1.11.3 An Investigation of Various Machine and Deep Learning Techniques Applied in Automatic Fear Level Detection and Acrophobia Virtual Therapy . . . . ..15 1.11.4 Acrophobia treatment with virtual reality –assisted acceptance and commitment therapy: two case reports . . . . . . . . . . . . . . 16 1.11.5 Electroencephalography correlates of fear of heights in a virtual reality environment . . . . . . .. 16 1.12 Conclusion . . . . . . . . . . . . . . . . . . . . . 17 2 Brain Computer Interface 18 2.1 Introduction . . . . . . . . . . . . . . . . . . . . . 18 2.2 Theoretical Background . . . . . . . . . . .. . . . . 18 2.2.1 Definition of Brain-Computer Interface . . . . . 18 2.2.2 Neurophysiological Foundations of BCI . . . . . . . . 19 2.2.3 Hybrid BCIs . . . . . . . . 28 2.3 Technical background . . . . . . . . . . . . . . . .. . . . . 29 2.3.1 Brain signals acquisition . . . . . .. . . 29 2.3.2 Preprocessing . . . . . . . . . . . . . . . . . . 31 2.3.3 Feature Extraction . . . . . . . . . . . . . . 32 2.3.4 Classification . . . . . . . . . . . 38 2.3.5 Feedback . . . . . . . . . . . . . . . . . . . . 39 2.4 Conclusion . . . . . . . . . . . . . . . . . . . 39 3 Design and Contribution 40 3.1 Introduction . . . . . . . . . . . . 40 3.2 General Design . . . . . . . . . . . . . .. . . . . 40 3.3 Detailed Design . . . . . . . . . . . . . . . . . . . 41 3.3.1 Virtual Reality Environment . . . . . . . . . . . . 41 3.3.2 Data Acquisition and Processing . . . . . . . 44 3.3.3 Acrophobia Level Recognition . . . . . . . . 45 3.3.4 Therapy Generation unit . . . . . . . . 53 3.4 Conclusion . . . . . . . . . . . . . . . . 56 v4 Implementation and Results 58 4.1 Introduction . . . . . . . . . . . . . . . . . 58 4.2 Development software and hardware . . . . . . . . . 58 4.2.1 VR Headset . . . . . . . . . . .. . 59 4.2.3 HPC using during VR session . . . . . . . . .. 60 4.2.4 Training Hardware . . . . . . . . . . . . . . . . . 60 4.2.5 Languages and Frameworks . . . . . . . . 60 4.3 System Overview . . . . . . . . . . . . . . . . . 64 4.3.1 Dataset . . . . . . . . . . . . . . . . . 64 4.3.2 Virtual Reality Results . . . . . . . . .. . . 67 4.3.3 Acrophobia Level Estimation . . . . . . . . . . 71 4.3.4 GAN architecture for the Next Scenario Generation . . . . .. 74 4.4 Conclusion . . . . . . . . . . . . . 80 Conclusion and future works 81 Bibliography 82 |
Disponibilité (1)
| Cote | Support | Localisation | Statut |
|---|---|---|---|
| Minf/805 | Mémoire master | bibliothèque sciences exactes | Consultable |
