Detecting cyberthreats in Metaverse learning platforms using an explainable DNN

Internet of Things, Год журнала: 2024, Номер 25, С. 101046 - 101046

Опубликована: Янв. 21, 2024

Язык: Английский

---

To pre-process or not to pre-process? On the role of EEG enhancement for cybersickness characterization and the importance of amplitude modulation features

Frontiers in Virtual Reality, Год журнала: 2025, Номер 6

Опубликована: Янв. 30, 2025

Virtual Reality (VR) has expanded beyond the entertainment field and become a valuable tool across different verticals, including healthcare, education, professional training, just to name few. Despite these advancements, widespread usage of VR systems is still limited, mostly due motion sickness symptoms, such as dizziness, nausea, headaches, which are collectively termed “cybersickness”. In this paper, we explore use electroencephalography (EEG) for real-time characterization cybersickness. particular, aim answer three research questions: (1) what neural patterns indicative cybersickness levels, (2) do EEG amplitude modulation features convey more important explainable patterns, (3) role does pre-processing play in overall characterization. Experimental results show that minimal retains artifacts may be useful detection (e.g., head eye movements), while advanced methods enable extraction interpretable help community gain additional insights on underpinnings Our experiments proposed comprise roughly 60% top-selected EEG-based detection.

Язык: Английский

---

“Are you feeling sick?” A systematic literature review of cybersickness in virtual reality

ACM Computing Surveys, Год журнала: 2024, Номер 56(11), С. 1 - 38

Опубликована: Июнь 3, 2024

Cybersickness (CS), also known as visually induced motion sickness (VIMS), is a condition that can affect individuals when they interact with virtual reality (VR) technology. This characterized by symptoms such nausea, dizziness, headaches, eye fatigue, and so on, be caused variety of factors. Finding feasible solution to reduce the impact CS extremely important it will greatly enhance overall user experience make VR more appealing wider range people. We have carefully compiled list 223 highly pertinent studies review current state research on most essential aspects CS. provided novel taxonomy encapsulates various measurement techniques found in literature. proposed set mitigation guidelines for both developers users. discussed CS-inducing factors tries capture same. Overall, our work provides comprehensive overview particular emphasis different strategies, identifies gaps literature, recommendations future field.

Язык: Английский

---

Electrogastrogram-based detection of cybersickness with the application of wavelet transformation and machine learning: A case study

Vojnotehnicki glasnik, Год журнала: 2025, Номер 73(1), С. 79 - 114

Опубликована: Янв. 1, 2025

Introduction/purpose: The application of virtual reality (VR) and simulation technologies in military training offers cost-effective versatile approach to enhancement. However, prevalence cybersickness (CS), characterized by symptoms such as nausea, limits their widespread use. Methods: This study introduces objective parameters for the detection CS using three-channel electrogastrogram (EGG) recording from one specific subject assesses independence linear correlation appropriate channel selection. paper employs a 3-level discrete wavelet transformation (DWT) on chosen identify key indicative gastric disturbances. Furthermore, investigates recovery following VR examines unsupervised machine learning (ML) segmenting EGG into baseline CS, utilizing significant features previously identified. Results discussion: analysis reveals no differences across channels moderate low between pairs. feature selection demonstrates that root mean square amplitude well maximum values power spectral density (PSD) calculated all DWT coefficients, are effective while dominant scale could not indicate any level decomposition. signs appear approximately 8 minutes after first experience supporting idea conducting multiple sessions same day i.e., intensive VR-based training. Conclusions: ML shows potential identifying CSaffected signal segments with extraction based DWT, offering novel enhancing prevention occurrence other VR-related environments.

Язык: Английский

---

Industrial Metaverse Towards Human-Centric Smart Manufacturing

Опубликована: Янв. 1, 2025

Язык: Английский

---

LiteVR: Interpretable and Lightweight Cybersickness Detection using Explainable AI

Опубликована: Март 1, 2023

Cybersickness is a common ailment associated with virtual reality (VR) user experiences. Several automated methods exist based on machine learning (ML) and deep (DL) to detect cyber-sickness. However, most of these cybersickness detection are perceived as computationally intensive black-box methods. Thus, those techniques neither trustworthy nor practical for deploying standalone energy-constrained VR head-mounted devices (HMDs). In this work, we present an explainable artificial intelligence (XAI)-based framework Lite detection, explaining the model's outcome, reducing feature dimensions, overall computational costs. First, develop three cybersick-ness DL models long-term short-term memory (LSTM), gated recurrent unit (GRU), multilayer perceptron (MLP). Then, employed post-hoc explanation, such SHapley Additive Explanations (SHAP), explain results extract dominant features cybersickness. Finally, retrain reduced number features. Our show that eye-tracking detection. Furthermore, XAI-based ranking dimensionality reduction, significantly reduce size by up 4.3×, training time 5.6×, its inference 3.8×, higher accuracy low regression error (i.e., Fast Motion Scale (FMS)). proposed lite LSTM model obtained 94% in classifying cyber-sickness regressing FMS 1–10) Root Mean Square Error (RMSE) 0.30, which outperforms state-of-the-art. can help researchers practitioners analyze, detect, deploy their DL-based HMDs.

Язык: Английский

---

Metaverse: Trend, emerging themes, and future directions

Transactions on Emerging Telecommunications Technologies, Год журнала: 2023, Номер 35(1)

Опубликована: Дек. 9, 2023

Abstract Metaverse is going to change human life in a profound way because it offers an opportunity merge our physical world with the digital/virtual worlds. Yet, how much effort has research community across and disciplines contributed what are emerging themes of metaverse? The study aims answer these important questions using bibliometric approach. Using search (“metaverse” OR “metaverses”) “Article Title, Abstract, Keywords” date range up 2022 on March 15, 2023 Scopus, 1031 journal articles, reviews, conference papers were identified. Among identified documents, 816 (i.e., around 80%) published 2022. Feiyue Wang Institute Automation, Chinese Academy Sciences was found be most productive author 17 metaverse publications, followed by Elif Ayiter Sabancı Üniversitesi 11 publications. (with its Automation) affiliation 29 China, United States, South Korea, Kingdom countries that over 58% Co‐occurrence keywords analysis revealed seven clusters emerged. included “artificial intelligence perception,” “metaverses blockchain,” “e‐learning students,” “metaverse, avatar immersive,” “virtual reality, virtual worlds, Second Life,” “three dimensional computer graphics deep learning,” “augmented mixed interaction.” Implications future directions given.

Язык: Английский

---

A Post-Stroke Rehabilitation System With Compensatory Movement Detection Using Virtual Reality and Electroencephalogram Technologies

IEEE Access, Год журнала: 2024, Номер 12, С. 61418 - 61432

Опубликована: Янв. 1, 2024

Stroke is a leading cause of global population mortality and disability, imposing burdens on patients caregivers, significantly affecting the quality life patients. Therefore, in this study, we aimed to explore application virtual reality technology physical therapy by using immersive interactive training designing rehabilitation modes for individual group settings. We also provide with stroke comprehensive home-based treatment plan, ultimately enhancing effectiveness. Patients can engage through system undergo functional, mirror, constraint-induced therapies tailored different task contents. Simultaneously, brain-computer interface technology, an emotion analysis mechanism was designed map patients' brainwave signal data onto two-dimensional space positive-negative valence arousal; approach enable remote therapists discern emotional states during process spaces, facilitating timely adjustments tasks. Moreover, prevent compromised effectiveness owing improper postures compensation, offers real-time identification recording, promptly issuing alerts when compensation occurs. The provides multiuser space, enabling corrections observations, offering program, thereby realizing localized aging care model.

Язык: Английский

---

Prediction and detection of virtual reality induced cybersickness: a spiking neural network approach using spatiotemporal EEG brain data and heart rate variability

Brain Informatics, Год журнала: 2023, Номер 10(1)

Опубликована: Июль 12, 2023

Virtual Reality (VR) allows users to interact with 3D immersive environments and has the potential be a key technology across many domain applications, including access future metaverse. Yet, consumer adoption of VR is limited by cybersickness (CS)-a debilitating sensation accompanied cluster symptoms, nausea, oculomotor issues dizziness. A leading problem lack automated objective tools predict or detect CS in individuals, which can then used for resistance training, timely warning systems clinical intervention. This paper explores spatiotemporal brain dynamics heart rate variability involved uses this information both episodes. The present study applies deep learning EEG spiking neural network (SNN) architecture prior using (85.9%, F7) it (76.6%, FP1, Cz). ECG-derived sympathetic (HRV) parameters prediction (74.2%) detection (72.6%) but at lower accuracy than EEG. Multimodal data fusion HRV does not change compared ECG alone. found that Cz (premotor supplementary motor cortex) O2 (primary visual are hubs functionally connected networks associated events susceptibility CS. F7 also suggested as area integrating implementing responses incongruent induce cybersickness. Consequently, Cz, presented here promising targets

Язык: Английский

---

Predicting VR cybersickness and its impact on visuomotor performance using head rotations and field (in)dependence

Frontiers in Virtual Reality, Год журнала: 2023, Номер 4

Опубликована: Ноя. 27, 2023

Introduction: This exploratory study aims to participate in the development of VR framework by focusing on issue cybersickness. The main objective is explore possibilities predicting cybersickness using i) field dependence-independence measures and ii) head rotations data through automatic analyses. second assess impact visuomotor performance. Methods: 40 participants completed a 13.5-min immersion first-person shooter game. Head were analyzed both their spatial (coefficients variations) temporal dimensions (detrended fluctuations analyses). Exploratory correlations, linear regressions clusters comparison (unsupervised machine learning) analyses performed explain Traditional human factors (sense presence, state flow, video game experience, age) also integrated. Results: Results suggest that measured before exposure ¼ variance cybersickness, while Disorientation scale Simulator Sickness Questionnaire predicts 16.3% In addition, during revealed two different participants, one them reporting more than other. Discussion: These results are discussed terms sensory integration diminution as an avoidance behavior negative symptoms. suggests measuring (Virtual) Rod Frame Test tracking internal sensors might serve powerful tools for actors.

Язык: Английский

---