Accurately modeling human intentions in pediatric cerebral palsy (CP) rehabilitation is essential for providing successful, adaptive therapy that responds to each child’s particular motor and cognitive characteristics. Conventional observation-based methods frequently fail to detect nuanced or unusual intention patterns, particularly in young children with intricate motor disorders. This study presents a theoretical framework that combines privacy-preserving federated learning (FL) with immersive extended reality (XR) technology to facilitate real-time, personalized intention recognition in therapeutic contexts. The system utilizes the immersive features of the Meta Quest Pro headset for interactive pediatric rehabilitation and the edge-processing capabilities of NVIDIA Jetson devices to do on-device inference and federated model updates without transferring sensitive patient information. The proposed architecture safeguards data privacy while facilitating decentralized model training in distant clinical settings. Our conceptual framework delineates multimodal data capture, federated aggregation procedures, adaptive XR feedback, and intention-aware therapeutic modifications—executed fully offline and under complete local control. This paper offers a scalable and ethically acceptable theoretical framework for revolutionizing pediatric rehabilitation using secure, intelligent, and immersive therapeutic technology, without necessitating implementation.

Advanced medical imaging has enhanced diagnostic accuracy and patient outcomes. Continued improvement means that the innovation presents significant medical benefits for health services, professionals and patients. However, access and adoption of the technology remain uneven due to the level of digital infrastructure and technical expertise required. The human and technical resources particularly impact rural and resource-constrained settings. These environments often face infrastructural limitations, unreliable connectivity, and restricted computational capacity, hindering equitable access to innovative technologies. In response, this research proposes a novel theoretical framework that integrates lightweight, quantization-enhanced deep learning with immersive offline virtual reality to generate high-fidelity tumor segmentation images tailored for low-resource contexts. This approach facilitates sporadic distant expert consultations, enhances local clinician training, and aligns medical technology deployment with environmental sustainability. While challenges remain in balancing accuracy, computational efficiency, patient acceptance, and regulatory compliance, this framework holds significant promise for advancing scalable, equitable healthcare delivery and diagnostic reliability in underserved settings.

Abstract Data augmentation is essential for enhancing computer vision performance, with the KeepOriginalAugment method standing out for intelligently incorporating salient and less prominent regions. Despite its success in image classification, its potential in addressing biases is unexplored. We introduce FaceKeepOriginalAugment, extending KeepOriginalAugment to address geographical, gender, and stereotypical biases in computer vision models. By balancing data diversity and information preservation, our approach enables models to leverage both salient and non-salient regions, fostering diversity and debiasing. We explore strategies for salient region placement and augmentation selection, quantifying diversity using Image Similarity Score (ISS) across datasets like FFHQ, WIKI, IMDB, LFW, and UTK Faces. We assess FaceKeepOriginalAugment in mitigating gender bias across CEO, Engineer, Nurse, and School Teacher datasets, using the Image-Image Association Score (IIAS) in CNNs and vision transformers (ViTs). Results show FaceKeepOriginalAugment effectively promotes fairness and inclusivity by reducing gender bias and enhancing fairness. Additionally, we introduce a Saliency-Based Diversity and Fairness Metric to quantify diversity and fairness while addressing data imbalance across datasets.

Deep learning-based models have revolutionized medical diagnostics by using Big Data to enhance disease diagnosis and clinical decision-making. However, their significant computational demands and opaque decision-making processes, often characterized as "black-box" systems, pose major challenges in time-critical and resource-constrained healthcare settings. To address these issues, this study explores the application of randomized machine learning models, specifically Extreme Learning Machines (ELMs) and Random Vector Functional Link (RVFL) networks, in medical diagnostics. These models introduce stochasticity into their training processes, reducing computational complexity and training times while maintaining accuracy. Furthermore, we integrate Explainable AI techniques namely Local Interpretable Model-agnostic Explanations (LIME) and Shapley Additive Explanations (SHAP) to explain the decision-making rationale of ELMs and RVFL. Performance evaluations on genitourinary cancers and coronary artery disease datasets demonstrate that RVFL outperforms traditional deep learning models, achieving superior accuracy of 88.29% with a computational overhead of 6.22 seconds for genitourinary cancers, and an accuracy of 81.64% with a computational time of 0.0308 seconds for coronary artery disease. This research highlights the potential of randomized models in enhancing efficiency and transparency in medical diagnosis, thereby accelerating better treatment outcomes and advocating for more accessible and interpretable AI solutions in healthcare.

The fairness of Artificial Intelligence (AI) has gained tremendous attention within the criminal justice system in recent years, mainly when predicting the risk of recidivism. The primary reason is attributed to evidence of bias towards demographic groups when deploying these AI systems. Many proposed fairness-improving techniques applied at each of the three phases of the fairness pipelines, pre-processing, in-processing and post-processing phases, are often ineffective in mitigating the bias and attaining high predictive accuracy. This paper proposes a novel approach by integrating existing fairness-improving techniques: Reweighing, Adversarial Learning, Disparate Impact Remover, Exponential Gradient Reduction, Reject Option-based Classification, and Equalized Odds optimization across the three fairness pipelines simultaneously. We evaluate the effect of combining these fairness-improving techniques on enhancing fairness and attaining accuracy. In addition, this study uses multi- and bi-objective optimization techniques to provide and to make well-informed decisions when predicting the risk of recidivism. Our analysis found that one of the most effective combinations (i.e., disparate impact remover, adversarial learning, and equalized odds optimization) demonstrates a substantial enhancement and balances achievement in fairness through various metrics without a notable compromise in accuracy.

This research presents an innovative framework that uses blockchain technology to improve tumor segmentation in medical imaging. The approach tackles issues related to data security, particularly when dealing with real private dataset, annotation accuracy, and collaboration. With the growing reliance of the medical industry on accurate tumor segmentation from medical images for cancer diagnosis and treatment, current methods are inadequate in maintaining data accuracy and promoting collaboration among experts across different countries. Our suggested approach utilizes blockchain technology to establish a decentralized, secure platform for the collaborative obtaining, annotation, and validation of medical images by data scientists, oncologists, and radiologists. Smart contracts streamline essential procedures such as verification of annotations, consensus among experts, and remuneration of contributors, guaranteeing the dependability and excellence of the data. Furthermore, the unchangeable record of transactions in the blockchain ensures a reliable basis for implementing artificial intelligence and machine learning algorithms. This improves the accuracy of segmenting data and allows for predictive modeling. This strategy not only improves the precision and effectiveness of tumor segmentation but also promotes a worldwide collaborative environment, which has the potential to revolutionize cancer diagnostics and treatment planning. Furthermore, it ensures the privacy and security of patient data.

Intelligent computational methods for economicsEconomics explores the behavior of people, companies, governments, and various decision-makers to explain how their decisions produce value and satisfy (or fail to satisfy) human needs and desires.Driven by advances in artificial intelligence (AI) and reinforced by the acumen of generated and collected open data, there is now a significant and growing field that utilizes the concept of a synthetic homo economicus, the mythical per-

Artificial Intelligence (AI) is applied in almost every public sector because of its positive impacts. However, AI’s ethical aspects and trustworthiness constitute a significant uproar and concern among different AI stakeholders due to AI’s adverse effect on users when the AI system lacks cautionary measures. AI is used in the criminal justice system for predicting recidivism risk. However, AI’s negative impact translates into bias and high incarceration towards a group of defendants in a population assessed for recidivism risk. This paper focuses on fairness as a requirement of a trustworthy AI framework previously proposed to ascertain the appropriate application of AI systems in predicting recidivism. This paper aims to raise awareness about the fairness of AI models and stimulate further research and deployment of efficient and effective exploitation of fair and trustworthy AI models in the criminal justice system when predicting recidivism. Fairness has been a significant concern for criminal justice system stakeholders and has received considerable attention with more theoretical and practical studies than other trustworthy AI requirements. Hence, this paper reviews state-of-the-art fairness, outlines valuable findings, and proposes future directions to achieve fair AI systems for predicting recidivism risk. In addition, this paper ensures mapping existing technical works in the literature to the fairness pipeline corresponding to the criminal justice system’s AI development phases.

New blockchain platforms are launching at a high cadence, each fighting for attention, adoption, and infrastructure resources. Several studies have measured the peer-to-peer (P2P) network decentralisation of Bitcoin and Ethereum (i.e., two of the largest used platforms). However, with the increasing demand for blockchain infrastructure, it is important to study node decentralisation across multiple blockchain networks, especially those containing a small number of nodes. In this paper, we propose NodeMaps, a data processing framework to capture, analyse, and visualise data from several popular P2P blockchain platforms, such as Cosmos, Stellar, Bitcoin, and Lightning Network. We compare and contrast the geographic distribution, the hosting provider diversity, and the software client variance in each of these platforms. Through our comparative analysis of node data, we found that Bitcoin and its Lightning Network Layer 2 protocol are widely decentralised P2P blockchain platforms, with the largest geographical reach and a high proportion of nodes operating on The Onion Router (TOR) privacy-focused network. Cosmos and Stellar blockchains have reduced node participation, with nodes predominantly operating in large cloud providers or well-known data centres.

Contact tracing apps used in tracing and mitigating the spread of COVID-19 have sparked discussions and controversies worldwide. The major concerns in relation to these apps are around privacy. Ireland was in general praised for the design of its COVID tracker app, and the transparency through which privacy issues were addressed. However, the ”voice” of the Irish public was not really heard or analysed. This study aimed to analyse the Irish public sentiment towards privacy and COVID tracker app. For this purpose we have conducted sentiment analysis on Twitter data collected from public Twitter accounts from Republic of Ireland. We collected COVID-19 related tweets generated in Ireland over a period of time from January 1, 2020 up to December 31, 2020 in order to perform sentiment analysis on this data set. Moreover, the study performed sentiment analysis on the feedback received from a national survey on privacy conducted in Republic of Ireland. The findings of the study reveal a significant criticism towards the app that relate to privacy concerns, but other aspects of the app as well. The findings also reveal some positive attitude towards the fight against COVID-19, but these are not necessarily related to the technological solutions employed for this purpose. The findings of the study contributed to the formulation of useful recommendations communicated to the relevant Irish actors.

Contact tracing apps used in tracing and mitigating the spread of COVID-19 have sparked discussions and controversies worldwide with major concerns around privacy. COVID Tracker app used in the Republic of Ireland was praised in general for the way it addressed privacy and was used as baseline for other contact tracing apps worldwide. The success of the app is dependent on the general public uptake, hence their voice and attitude is the one that really matters. This paper focuses on developing a survey and the methods aiming to examine the attitudes toward privacy during COVID-19 of the general public in the Republic of Ireland and their impact on the uptake of the COVID tracker app. Various privacy models are used and health belief model as well in this purpose. A pilot study with 286 participants show a change in attitude towards privacy during COVID-19 pandemic, with more people willing to share their data in the interest of saving lives. However, privacy attitudes are shown to have impacted the adoption of the app in Ireland.

Simulation can be a powerful technique for evaluating the performance of large-scale cloud computing services in a relatively low cost, low risk and time-sensitive manner. Large-scale data indexing, distribution and management is complex to analyse in a timely manner. In this paper, we extend the CloudSim cloud simulation framework to model and simulate a distributed search engine architecture and its workload characteristics. To test the simulation framework, we develop a model based on a real-world ElasticSearch deployment on Linknovate.com. An experimental evaluation of the framework, comparing simulated and actual query response time, precision and resource utilisation, suggests that the proposed framework is capable of predicting performance at different scales in a precise, accurate and efficient manner. The results can assist ElasticSearch users to manage their scalability and infrastructure requirements.

Clustering techniques are very attractive for extracting and identifying patterns in datasets. However, their application to very large spatial datasets presents numerous challenges such as high-dimensionality data, heterogeneity, and high complexity of some algorithms. For instance, some algorithms may have linear complexity but they require the domain knowledge in order to determine their input parameters. Distributed clustering techniques constitute a very good alternative to the big data challenges (e.g.,Volume, Variety, Veracity, and Velocity). Usually these techniques consist of two phases. The first phase generates local models or patterns and the second one tends to aggregate the local results to obtain global models. While the first phase can be executed in parallel on each site and, therefore, efficient, the aggregation phase is complex, time consuming and may produce incorrect and ambiguous global clusters and therefore incorrect models. In this paper we propose a new distributed clustering approach to deal efficiently with both phases, generation of local results and generation of global models by aggregation. For the first phase, our approach is capable of analysing the datasets located in each site using different clustering techniques. The aggregation phase is designed in such a way that the final clusters are compact and accurate while the overall process is efficient in time and memory allocation. For the evaluation, we use two well-known clustering algorithms, K-Means and DBSCAN. One of the key outputs of this distributed clustering technique is that the number of global clusters is dynamic, no need to be fixed in advance. Experimental results show that the approach is scalable and produces high quality results.