Mohd Fairuz Shiratuddin

Gamification plays a pivotal role in enhancing user engagement in the Metaverse, particularly among Generation Z users who value autonomy, immersion, and identity expression. However, current research lacks a cohesive framework tailored to designing gamified social experiences in immersive virtual environments. This study presents a framework-oriented systematic literature review, guided by PRISMA 2020 and SPIDER, to investigate how gamification is applied in the Metaverse and how it aligns with the behavioral needs of Gen Z. From 792 screened studies, seventeen high-quality papers were synthesized to identify core gamification mechanics, including avatars, XR affordances, and identity-driven engagement strategies. Building on these insights, we propose the Affordance-Driven Gamification Framework (ADGF), a conceptual model for designing socially immersive experiences, along with a five-step design process to support its real-world application. Our contributions include a critical synthesis of existing strategies, Gen Z-specific design considerations, and a dual-framework approach to guide researchers and practitioners in developing emotionally engaging and socially dynamic Metaverse experiences.

Objectives
The objective of this research was to investigate if the use of virtual reality, increasingly utilised within intensive care medicine due to its demonstrated benefits in improving pain and anxiety, has been reported to result in seizures.

Review method used
A rapid systematic review and synthesis of qualitative and quantitative data was performed.

Data sources
Five databases (PubMed, Scopus, EMBASE, PsycInfo, and CINAHL) were systematically searched. An additional gray literature search was also conducted. Articles were restricted to those published on or after January 1st, 2014.

Review methods
The number of participants, virtual reality sessions, and length of sessions was undertaken. Subgroup analysis was undertaken for both adult and paediatric patient populations. An additional subgroup analysis was undertaken on articles which did not exclude individuals with a history of epilepsy. A tailored risk-of-bias assessment was conducted.
Results
Of the 563 articles identified through database and gray literature searching, 27 articles met inclusion criteria. A total of 886 patients have been reported within the literature with a combined 1843 virtual reality sessions, totalling more than 614.64 h of virtual reality. No seizures have been reported within intensive care patients receiving virtual reality interventions.

Conclusions
Historically, individuals with a history of epilepsy and photosensitivity have been commonly excluded from interventions and clinical trials involving virtual reality. The results of this systematic review demonstrate that the risk of virtual reality is minimal when utilised appropriately. A history of photosensitivity or epilepsy should not constitute an absolute contraindication for the use of virtual reality. Instead, clinicians should utilise clinical judgement when evaluating a patient's risk and ensure that appropriate visual experiences are utilised which do not unnecessarily strobe the patient.

The main contribution of this paper is to introduce a framework for integrating Machine Learning (ML), Human, and Virtual Reality (VR) into one platform to promote a collaborative visualisation environment that can assist in better analysis and improve the human-machine teaming capability. This platform was demonstrated using a case study in ana-lysing road traffic conditions. The ‘Ab-normal Machine Learning Road Traffic Detection in VR (AbnMLRTD-VR)’ prototype system was developed to assist the human analyst. The proposed system has two main integrative components: a data-driven ML model and a 3D real-time visualisation in a VR environment. An unsupervised ML model was built using real traffic data. The AbnMLRTD-VR system highlights the outliers in the road sections in actual road contexts of a road traffic network. This gives the human analyst a 3D real-time immersive visualisation in a VR environment to evaluate road conditions. The AbnMLRTD-VR system demonstrated that it could help minimise the need for human pre-labelling of the data. It enables the visualisation of the road traffic conditions more meaningfully and to understand the context of the road traffic conditions of road sections at any given time.

To improve users’ tendency towards shopping in Virtual Reality (VR), en-hancing the User Experience (UX) of the VR shopping environments is of primary importance. Product viewability, reachability, and personalisation are some of the primary UX factors in a shopping environment. This paper proposes and discusses three factors for a Personalised Adaptive Aisle (PAA) in a VR shopping environment to improve the shopping experience. They are 1) Shelf placement for viewability and reachability, 2) User view-point in VR, and 3) Personalised Product placement.

Deep learning (DL) has achieved superior classification in many applications due to its capability of extracting features from the data. However, the success of DL comes with the tradeoff of possible overfitting. The bias towards the data it has seen during the training process leads to poor generalisation. One way of solving this issue is by having enough training data so that the classifier is invariant to many data patterns. In the literature, data augmentation has been used as a type of regularisation method to reduce the chance for the model to overfit. However, most of the relevant works focus on image, sound or text data. There is not much work on numerical data augmentation, although many real-world problems deal with numerical data. In this paper, we propose using a technique based on Fuzzy C-Means clustering and fuzzy membership grades. Fuzzy-related techniques are used to address the variance problem by generating new data items based on fuzzy numbers and each data item's belongings to different fuzzy clusters. This data augmentation technique is used to improve the generalisation of a Deep Neural Network that is suitable for numerical data. By combining the proposed fuzzy data augmentation technique with the Dropout regularisation technique, we manage to balance the classification model's bias-variance tradeoff. Our proposed technique is evaluated using four popular data sets and is shown to provide better regularisation and higher classification accuracy compared with popular regularisation approaches.

Engagement with upper limb rehabilitation post-stroke can improve rehabilitation outcomes. Virtual Reality can be used to make rehabilitation more engaging. In this paper, we propose a multiple case study to determine: (1) whether game design principles (identified in an earlier study as being likely to engage) actually do engage, in practice, a sample of stroke survivors with a Desktop Virtual Reality-based Serious Game designed for upper limb rehabilitation; and (2) what game design factors support the existence of these principles in the game. In this study, we considered 15 principles: awareness , feedback , interactivity , flow , challenge , attention , interest , involvement , psychological absorption , motivation , effort , clear instructions , usability , purpose , and a first-person view . Four stroke survivors used, for a period of 12 weeks, a Virtual Reality-based upper limb rehabilitation system called the Neuromender Rehabilitation System. The stroke survivors were then asked how well each of the 15 principles was supported by the Neuromender Rehabilitation System and how much they felt each principle supported their engagement with the system. All the 15 tested principles had good or reasonable support from the participants as being engaging. Use of feedback was emphasised as an important design factor for supporting the design principles, but there was otherwise little agreement in important design factors among the participants. This indicates that more personalised experiences may be necessary for optimised engagement. The insight gained can be used to inform the design of a larger scale statistical study into what engages stroke survivors with Desktop Virtual Reality-based upper limb rehabilitation.

Class imbalance is a serious issue in classification as a traditional classifier is generally biased towards the majority class. The accuracy of the classifier could be further impacted in cases where additionally to the class imbalance, there are overlapped data instances. Further, data sparsity has shown to be a possible issue that may lead to non- invariance and poor generalisation. Data augmentation is a technique that can handle the generalisation issue and improve the regularisation of the Deep Neural Network (DNN). A method to handle both class overlap and class imbalance while also incorporating regularisation is proposed in this paper. In our work, the imbalanced dataset is balanced using SMOTETomek, and then the non-categorical attributes are fuzzified. The purpose of fuzzifying the attributes is to handle the overlapping in the data and provide some form of data augmentation that can be used as a regularisation technique. Therefore, in this paper, the invariance is achieved as the augmented data are generated based on the fuzzy concept. The balanced augmented dataset is then trained using a DNN classifier. The datasets used in the experiments were selected from UCI and KEEL data repositories. The experiments show that the proposed Fuzzy data augmentation for handling overlapped and imbalanced data can address the overlapped and imbalanced data issues, and provide regularisation using data augmentation for numerical data to improve the performance of a DNN classifier.

Generating textual descriptions of images has been an important topic in computer vision and natural language processing. A number of techniques based on deep learning have been proposed on this topic. These techniques use human-annotated images for training and testing the models. These models require a large number of training data to perform at their full potential. Collecting human generated images with associative captions is expensive and time-consuming. In this paper, we propose an image captioning method that uses both real and synthetic data for training and testing the model. We use a Generative Adversarial Network (GAN) based text to image generator to generate synthetic images. We use an attention-based image captioning method trained on both real and synthetic images to generate the captions. We demonstrate the results of our models using both qualitative and quantitative analysis on popularly used evaluation metrics. We show that our experimental results achieve two fold benefits of our proposed work: i) it demonstrates the effectiveness of image captioning for synthetic images, and ii) it further improves the quality of the generated captions for real images, understandably because we use additional images for training.

The performance of classification algorithms with highly imbalanced streaming data depends upon efficient balancing strategy. Some techniques of balancing strategy have been applied using static batch data to resolve the class imbalance problem, which is difficult if applied for massive data streams. In this paper, a new Piece-Wise Incremental Data re-Balancing (PWIDB) framework is proposed. The PWIDB framework combines automated balancing techniques using Racing Algorithm (RA) and incremental rebalancing technique. RA is an active learning approach capable of classifying imbalanced data and can provide a way to select an appropriate re-balancing technique with imbalanced data. In this paper, we have extended the capability of RA for handling imbalanced data streams in the proposed PWIDB framework. The PWIDB accumulates previous knowledge with increments of re-balanced data and captures the concept of the imbalanced instances. The PWIDB is an incremental streaming batch framework, which is suitable for learning with streaming imbalanced data. We compared the performance of PWIDB with a well-known FLORA technique. Experimental results show that the PWIDB framework exhibits an improved and stable performance compared to FLORA and accumulative re-balancing techniques.

In spectator crowd images, the high number of people, small size and occlusion of body parts, make the body detection task challenging. Due to the similarity in facial features of different people, the variance in head features is less compared to the variation in the body features. Similarly, the visibility of the head in a crowd is more, compared to the visibility of the body. Therefore, the detection of only the head is more successful than the detection of the full body. We show that there exists a relation between head size and location, and the body size and location in the image. Therefore, head size and location can be leveraged to detect full bodies. This paper suggests that due to lack of visibility, more variance in body features, and lack of available training data of occluded bodies, full bodies should not be detected directly in occluded scenes. The proposed strategy is to detect full bodies using information extracted from head detection. Additionally, body detection technique should not be affected by the level of occlusion. Therefore, we propose to use only color matching for body detection. It does not require any explicit training data like CNN based body detection. To evaluate the effectiveness of this strategy, experiments are performed using the S-HOCK spectator crowd dataset. Using partial ground truth head information as the input, full bodies in a dense crowd is detected. Experimental results show that our technique using only head detection and color matching can detect occluded full bodies in a spectator crowd successfully.