Towards Machine Learning Driven Self-guided Virtual Reality Exposure Therapy Based on Arousal State Detection from Multimodal Data

Rahman, Muhammad Arifur; Brown, David J.; Shopland, Nicholas; Harris, Matthew C.; Turabee, Zakia Batool; Heym, Nadja; Sumich, Alexander; Standen, Bradley; Downes, David; Xing, Yangang; Thomas, Carolyn; Haddick, Sean; Premkumar, Preethi; Nastase, Simona; Burton, Audrey; Lewis, James E.; Mahmud, Mufti

doi:10.1007/978-3-031-15037-1_17

Cited by 20 publications

(5 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Studies have been performed where VR has simulated an MRI experience ( [45]), VR has been used to prepare children for MRI ( [46]), and studies have used VR during hemoencephalographic ( [10]) and MRI sessions ( [47]). Recent studies have explored the use of fMRI VR-like stimuli to prepare PTSD patients for VR training ( [48]), machine learning of biofeedback to optimize VR exposure therapy ( [49]), the combination of biofeedback, VR, and mobile technology to enable home training through sham-feedback ( [50]), and NIRS neurofeedback for at-home use by patients ( [51]). To the best of the authors' knowledge, this is the first protocol aiming to explore how targeted use of contextual memory/VR immersion, and home training with sham-feedback may enhance subsequent real-time fMRI neurofeedback sessions.…”

Section: Introductionmentioning

confidence: 99%

Exploring protocol development: Implementing systematic contextual memory to enhance real-time fMRI neurofeedback

Fagerland,

Berntsen,

Fredriksen

et al. 2024

Journal of Electrical Bioimpedance

View full text Add to dashboard Cite

Objective The goal of this study was to explore the development and implementation of a protocol for real-time fMRI neurofeedback (rtfMRI-nf) and to assess the potential for enhancing the selective brain activation using stimuli from Virtual Reality (VR). In this study we focused on two specific brain regions, supplementary motor area (SMA) and right inferior frontal gyrus (rIFG). Publications by other study groups have suggested impaired function in these specific brain regions in patients with the diagnoses Attention Deficit Hyperactivity Disorder (ADHD) and Tourette’s Syndrome (TS). This study explored the development of a protocol to investigate if attention and contextual memory may be used to systematically strengthen the procedure of rtfMRI-nf. Methods We used open-science software and platforms for rtfMRI-nf and for developing a simulated repetition of the rtfMRI-nf brain training in VR. We conducted seven exploratory tests in which we updated the protocol at each step. During rtfMRI-nf, MRI images are analyzed live while a person is undergoing an MRI scan, and the results are simultaneously shown to the person in the MRI-scanner. By focusing the analysis on specific regions of the brain, this procedure can be used to help the person strengthen conscious control of these regions. The VR simulation of the same experience involved a walk through the hospital toward the MRI scanner where the training sessions were conducted, as well as a subsequent simulated repetition of the MRI training. The VR simulation was a 2D projection of the experience. The seven exploratory tests involved 19 volunteers. Through this exploration, methods for aiming within the brain (e.g. masks/algorithms for coordinate-system control) and calculations for the analyses (e.g. calculations based on connectivity versus activity) were updated by the project team throughout the project. The final procedure involved three initial rounds of rtfMRI-nf for learning brain strategies. Then, the volunteers were provided with VR headsets and given instructions for one week of use. Afterward, a new session with three rounds of rtfMRI-nf was conducted. Results Through our exploration of the indirect effect parameters – brain region activity (directed oxygenated blood flow), connectivity (degree of correlated activity in different regions), and neurofeedback score – the volunteers tended to increase activity in the reinforced brain regions through our seven tests. Updates of procedures and analyses were always conducted between pilots, and never within. The VR simulated repetition was tested in pilot 7, but the role of the VR contribution in this setting is unclear due to underpowered testing. Conclusion This proof-of-concept protocol implies how rtfMRI-nf may be used to selectively train two brain regions (SMA and rIFG). The method may likely be adapted to train any given region in the brain, but readers are advised to update and adapt the procedure to experimental needs.

show abstract

Section: Introductionmentioning

confidence: 99%

Exploring protocol development: Implementing systematic contextual memory to enhance real-time fMRI neurofeedback

Fagerland,

Berntsen,

Fredriksen

et al. 2024

Journal of Electrical Bioimpedance

View full text Add to dashboard Cite

show abstract

“…In recent years artificial intelligence (AI) has been applied in diverse problem domains to solve various challenging problems including student engagement [10], virtual reality exposure therapy [11], text classification [12,13,14,15], cyber security [16,17,18,19], neurological disease detection [20,21,22] and management [23,24,25,26,27,28], elderly care [29,30], biological data mining [31,32], fighting pandemic [33,34,35,36,37,38,39], and healthcare service delivery [40,41,42]. In particular, convolutional neural network (CNN) has recently accomplished remarkable achievements in Pathology Informatics [43,44,45] and Bioinformatics [46,47,48], suggesting that CNN has the ability to solve these challenges.…”

Section: Introductionmentioning

confidence: 99%

Detection of Osteosarcoma using Deep Feature Extraction with Federated Learning and MLP from Histopathological Images

Chowdhury,

Mahmud,

Elahe

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

Background: The most common type of cancerous bone tumour is Osteosarcoma. Microscopic analysis of coloured slides is mostly used for Osteosarcoma diagnosis. As Osteosarcoma is highly heterogeneous, pathologists struggle to get a consensus diagnosis. Moreover, manual analysis of microscopic images is laborious, time-consuming, and vulnerable to bias. Method: In this research, a framework that extracts features based on CNN is developed to predict Osteosarcoma to address previous studies' performance gaps and shortcomings. The dataset used for this experiment is obtained from pathology archives at Children's Medical Center in Dallas, which includes 1144 histology glass slides. First, data augmentation is used to extend the dataset size. This study found Blur, Contrast, and Random augmentation perform well for the Osteosarcoma dataset. Next, we primarily considered six pre-trained transfer learning (TL) models and modified the models to extract features. Then, we examined feature selectors like Genetic Algorithm, Principal Component Analysis, Recursive Feature Elimination and Baruto and found PCA selects the optimal feature set. Finally, we fed the selected features into a fine-tuned MLP called Grid search MLP (GsMLP) and utilised federated learning (FL) to ensure the diversity and privacy of the dataset. Results: Experiments have revealed that the modified Xception model performs better in extracting features for Osteosarcoma prediction. Also, the proposed framework performed better than state-of-the-art techniques such as Convolutional Neural Network, VGG19, Random Forest, and AdaBoost with 98.20% accuracy and 97.5% Receiver Operating Characteristic - Area Under the Curve.

show abstract

“…Here, the correct detection of physiological states through robust models for the effective management of anxiety-induced arousal or stress is pivotal to facilitating intervention and enhancing psychological health and well-being. We have presented our first abstract concept Machine Learning Driven Self-guided Virtual Reality Exposure Therapy Based on Arousal State Detection from Multimodal Data in [9]. Then we started implementation, and here in this paper, we have added the concept of Biofeedback as a form of variation of heart rate and laterality index using EEG data and synthesised heart rate collected by emotive EPOC flex [?…”

Section: Introductionmentioning

confidence: 99%

Biofeedback Towards Machine Learning Driven Self-guided Virtual Reality Exposure Therapy Based on Arousal State Detection from Multimodal Data

Rahman

Brown

Mahmud

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Virtual-reality exposure therapy (VRET) is a novel intervention technique that allows individuals to experience anxiety-evoking stimuli in a safe environment, recognise specific triggers and gradually increase their exposure to perceived threats. Public-speaking anxiety (PSA) is a prevalent form of social anxiety, characterised by stressful arousal and anxiety generated when presenting to an audience. In self-guided VRET, participants can gradually increase their tolerance to exposure and reduce anxiety-induced arousal and PSA over time. However, creating such a VR environment and determining physiological indices of anxiety-induced arousal or distress is an open challenge. Environment modelling, character creation and animation, psychological state determination and the use of machine learning models for anxiety or stress detection are equally important, and multi-disciplinary expertise is required.In this work, we have explored a series of machine learning models with publicly available data sets (using electroencephalogram and heart rate variability) to predict arousal states. If we can detect anxiety-induced arousal, we can trigger calming activities to allow individuals to cope with and overcome distress. Here, we discuss the means of effective selection of machine learning models and parameters in arousal detection. We propose a pipeline to overcome the model selection problem with different parameter settings in the context of Virtual Reality Exposure Therapy. This pipeline can be extended to other domains of interest where arousal detection is crucial. Finally, we have implemented a biofeedback framework for VRET where we successfully provided feedback as a form of heart rate and brain laterality index from our acquired multimodal data for psychological intervention to overcome anxiety.

show abstract

Towards Machine Learning Driven Self-guided Virtual Reality Exposure Therapy Based on Arousal State Detection from Multimodal Data

Cited by 20 publications

References 35 publications

Exploring protocol development: Implementing systematic contextual memory to enhance real-time fMRI neurofeedback

Exploring protocol development: Implementing systematic contextual memory to enhance real-time fMRI neurofeedback

Detection of Osteosarcoma using Deep Feature Extraction with Federated Learning and MLP from Histopathological Images

Biofeedback Towards Machine Learning Driven Self-guided Virtual Reality Exposure Therapy Based on Arousal State Detection from Multimodal Data

Contact Info

Product

Resources

About