10th International Congress on Information and Communication Technology in concurrent with ICT Excellence Awards (ICICT 2025) will be held at London, United Kingdom | February 18 - 21 2025.
Sign up or log in to bookmark your favorites and sync them to your phone or calendar.
Authors - Reena (Mahapatra) Lenka, Smita Mehendale Abstract - This research paper examines this combination to understand better how Artificial Intelligence (AI) and Extended Reality (XR) might work together to change human experiences and capacities. Enhancing immersive environments primarily depends on artificial intelligence (AI), miming human cognitive capabilities. The study conducts a thorough literature review to comprehend the AI-XR synergy's goals, uses, constraints, and viewpoints. It emphasises how AI may complement human labour and how XR can produce multi-dimensional experiences, using examples from the aerospace, construction, and healthcare sectors. The paper describes the influence of these technologies on the nature of work in the future. Also, it focuses on the necessity for companies to create strategies that take advantage of both possibilities and problems. To effectively recruit and develop the talents necessary to merge human and machine efforts in modern workplaces, the H.R. role is growing.
Authors - Byron Albuja-Sanchez, Jeniffer Flores-Toala, Arcesio Bustos-Gaibor, Sandra Arias-Villon Abstract - The problem with predicting a possible risk of fetal death is that it depends on several aspects, not only medical but also economic and social aspects of the pregnant mothers, which makes an early response to this problem very difficult. This work seeks to apply classification algorithms to detect the risk of fetal death based on socioeconomic and demographic data of pregnant mothers in Ecuador, using datasets from 2000 to 2021. Trained algorithms include decision trees, random forests, neural networks, bagging classifiers, k nearest neighbors, and naive bayes bernoulli. As cases of fetal death are very rare, over-sampling and undersampling techniques were applied to train the algorithms. The performance comparison of the trained algorithms was carried out with their respective confusion matrices. The best performance was obtained by the algorithms trained with undersampling and of all of them the performance of the neural network stood out. The best performance of the neural network was associated with its nature of classifying by assigning weights to each input parameter.
Authors - Christos Chronis, Iraklis Varlamis, Konstantinos Tserpes, George Dimitrakopoulos, Faycal Bensaali Abstract - This paper introduces an innovative architecture for deploying AI models in edge and cloud environments, leveraging Federated Learning and RISC-V processors for privacy and real-time inference. It addresses the constraints of edge devices like Raspberry Pi and Jetson Nano by training models locally and aggregating results in the cloud to mitigate overfitting and catastrophic forgetting. RISC-V processors enable high-speed inference at the edge. Applications include energy consumption monitoring with LSTM models and recommendations via collaborative filtering, and multi-robot human collaboration using CNN and YOLO models. Model compression and partitioning optimize performance on RISC-V, with experiments demonstrating scalability and responsiveness under varying computational demands.
Authors - Joyce Wong Ching Yan, Davy Tsz Kit Ng Abstract - This chapter investigates the effects of generative artificial intelligence and digital transformation on K-16 school leaders in the post-pandemic period. It describes the challenges those leaders face in the technology integration processes and recommends developing AI competencies, interdisciplinary curricula and relevant leadership skills. The shift to learning over the Internet as a consequence of the actions that were taken towards the prevention of the spread of covid 19 brought many positives but many education managers continue to face the challenges of educational technology integration. Such factors include the influence of wonderful ideas and the provision of working rooms that encourage teacher collaboration. The chapter proposes plans for creating professional development furthering teachers’ modernization of their digital knowledge. It also addresses the aspect of the efficient digital design that helps implement contemporary curricular programs. As soon as they pay attention to these most important items, their schools will be more able to meet the needs of the digital world market, and consequently enhance the students’ performance. This chapter adds further discussion of initiatives and projects on school leadership and technology communication integration. It describes specific techniques relevant to today's education and the issues that they face as the environment changes rapidly.
Authors - Mohammad Hamzehloui, Ardavan Ashabi Abstract - Microservices have emerged as a preferred architectural style for developing scalable and resilient applications, especially within cloud environments. This approach offers significant advantages over traditional monolithic architectures, such as enhanced scalability, flexibility, and fault isolation. However, these benefits come with substantial operational costs. Running microservices on cloud platforms incurs high expenses due to the need for extensive monitoring, complex service management, and dynamic resource allocation. Industry solutions have primarily focused on monitoring and management, leaving a gap in comprehensive strategies for cost reduction through optimization and resource management. This study aims to identify and analyze the primary cost drivers of running microservices and assess their individual impacts. By providing a detailed analysis, this research enhances the understanding of cost factors, aiding in the cost management and optimization of cloud-based microservices. This knowledge helps businesses make informed decisions to minimize expenses while maximizing the benefits of cloud adoption. Key cost drivers identified include virtualization mechanisms, scaling solutions, microservice architectures, API designs. Microservices vary significantly in terms of performance and resource consumption depending on their design and architecture. However, by following certain best practices, it is possible to reduce the overall running costs of microservices by minimizing resource consumption.
Authors - Rachel Roux, Sonia Yassa, Olivier Romain Abstract - Cloud computing is widely used to collect data from various devices, which must be processed quickly. To manage this growing data, fog computing helps reduce delay and processing costs by assigning tasks to suitable devices. This article presents an adapted binary monarch butterfly algorithm for task offloading in a fog-cloud environment. This metaheuristic directly constructs a Pareto front, offering a solution space representation. Two versions are examined: one using random search and the other a deterministic search with crowding distance. Simulations on tasks from 40 to 500 show that the binary Monarch Butterfly algorithm can outperform state-of-the-art algorithms for cost optimization while balancing delay.
Authors - Leo Thomas Ramos, Angel D. Sappa Abstract - This work explores the integration of a Channel Attention (CA) module into the ConvNeXt architecture to improve performance in scene classification tasks. Using the UC Merced dataset, experiments were conducted with two data splits: 50% and 20% for training. Models were trained for up to 20 epochs, limiting the training process to assess which models could extract the most relevant features efficiently under constrained conditions. The ConvNeXt architecture was modified by incorporating a Squeeze-and-Excitation block, aiming to enhance the importance of each feature channel. ConvNeXt models with CA showed strong results, achieving the highest performance in the experiments conducted. ConvNeXt large with CA reached 90% accuracy and 89.75% F1-score with 50% of the training data, while ConvNeXt base with CA achieved 77.14% accuracy and 75.23% F1-score when trained with only 20% of the data. These models consistently outperformed their standard counterparts, as well as other architectures like ResNet and Swin Transformer, achieving improvements of up to 9.60% in accuracy, highlighting the effectiveness of CA in boosting performance, particularly in scenarios with limited data.
Authors - Hamada R. H. Al-Absi, Devi G. Kurup, Amina Daoud, Jens Schneider, Wajdi Zaghouani, Saeed Mohd H. M. Al Marri, Younss Ait Mou Abstract - This study integrates traditional Science of Hadith literature —documenting the sayings, actions, and approvals of the Prophet Muhammad (PBUH) with modern digital tools to analyze the geographic and temporal data of Hadith narrators. Using the Kaggle Hadith Narrators dataset, we apply Kernel Density Estimation (KDE) to map the spatial distribution of narrators’ birthplaces, places of stay, and death locations across generations, revealing key geographical hubs of Hadith transmission, such as Medina, Baghdad, and Nishapur. By examining narrators’ timelines and locations, we illustrate movement patterns and meeting points over time, providing insights into the spread of Hadith across the Islamic world during early Islamic history. To our knowledge, this research is the first systematic attempt to analyze Hadith transmission using geo-spatial and temporal methods, offering a novel perspective on the geographic and intellectual dynamics of early Islamic scholarship.
Authors - Francisco Seipel-Soubrier, Jonathan Cyriax Brast, Eicke Godehardt, Jorg Schafer Abstract - We propose an architecture of a proof-of-concept for automated video summarization and evaluate its performance, addressing the challenges posed by the increasing prevalence of video content. The research focuses on creating a multi-modal approach that integrates audio and visual analysis techniques to generate comprehensive video descriptions. Evaluation of the system across various video genres revealed that while video-based large language models show improvements over image-only models, they still struggle to capture nuanced visual narratives, resulting in generalized output for videos without a strong speech based narrative. The multi-modal approach demonstrated the ability to generate useful short summaries for most video types, but especially in speech-heavy videos offers minimal advantages over speech-only processing. The generation of textual alternatives and descriptive transcripts showed promise. While primarily stable for speech-heavy videos, future investigation into refinement techniques and potential advancements in video-based large language models holds promise for improved performance in the future.
Authors - A.G.H.R. Godage, H.R.O.E. Dayaratna Abstract - This study explores the implementation of edge computing for semi-automated vehicle systems in urban environments, leveraging modern wireless technologies such as 5G for efficient data transmission and processing. The proposed framework integrates a vehicle-mounted camera, an edge server, and deep learning models to identify critical objects, such as pedestrians and traffic signals, and predict vehicle speeds for the subsequent 30 seconds. By offloading computationally intensive tasks to an edge server, the system reduces the vehicle’s processing load and energy consumption, while embedded offline models ensure operational continuity during network disruptions. The research focuses on optimizing image compression techniques to balance bandwidth usage, transmission speed, and prediction accuracy. Comprehensive experiments were conducted using the Zenseact Open Dataset, a new dataset published in 2023, which has not yet been widely utilized in the domain of semi-automated vehicle systems, particularly for tasks such as predictive speed modeling. The study evaluates key metrics, including bandwidth requirements, round-trip time (RTT), and the accuracy of various machine learning and neural network models. The results demonstrate that selective image compression significantly reduces transmission times and overall RTT without compromising prediction quality, enabling faster and more reliable vehicle responses. This work contributes to the development of scalable, energy-efficient solutions for urban public transport systems. It highlights the potential of integrating edge AI frameworks to enhance driving safety and efficiency while addressing critical challenges such as data transmission constraints, model latency, and resource optimization. Future directions include extending the framework to incorporate multi modality, broader datasets, and advanced communication protocols for improved scalability and robustness.
Authors - Haesol Kim, Eunjae Kim, Sou Hyun Jang, Eun Kyoung Shin Abstract - This study empirically examined historical trajectories of the semantic landscape of legal conflicts over medical decision making. We unveiled the lexical structures of lawsuit verdicts, tracing how the core concepts of shared decision making (SDM)-duty of care, duty to explain, self-determination-have developed and been contextualized in legal discourses. We retrieved publicly available court verdicts using the search keyword ‘patient’ and screened them for relevance to doctor-patient communications. The final corpus comprised 251 South Korean verdicts issued between 1974 and 2023. We analyzed the verdicts using neural topic modeling and semantic network analysis. Our study showed that topic diversity has expanded over time, indicating increased complexity of semantic structures regarding medical decision-making conflicts. We also found two dominant topics: disputes over healthcare providers’ liability and disputes over the compensation for medical malpractice. The results of semantic network analysis showed that the rhetorics of patients’ right to medical self-determination are not closely tied to the professional responsibility to explain and care. The decoupled semantic relationships of patients’ right and health professionals’ duties revealed the barriers of SDM implementations.
Authors - Given Sichilima, Jackson Phiri Abstract - the health and productivity of poultry farms are significantly impacted by the timely detection of diseases within chicken houses. Manual disease monitoring in poultry is laborious and prone to errors, underscoring the need for sustainable, efficient, reliable, and cost-effective farming practices. The adoption of advanced technologies, such as artificial intelligence (AI), is essential to address this need. Smart farming solutions, particularly machine learning, have proven to be effective predictive analytical tools for large volumes of data, finding applications in various domains including medicine, finance, and sports, and now increasingly in agriculture. Poultry diseases, including coccidiosis, can significantly impact chicken productivity if not identified early. Machine learning and deep learning algorithms can facilitate earlier detection of these diseases. This study introduces a framework that employs a Convolutional Neural Network (CNN) to classify poultry diseases by examining fecal images to distinguish between healthy and unhealthy samples. Unhealthy fecal images may indicate the presence of disease. An image classification dataset was utilized to train the model, which achieved an accuracy of 84.99% on the training set and 90.05% on the testing set. The evaluation indicated that this model was the most effective for classifying chicken diseases. This research underscores the benefits of automated disease detection within smart farming practices in Zambia.
Authors - Ruqaya Majeed Kareem, Mohammed Kh. Al-Nussairi Abstract - Since the establishment of microgrids, the frequency stability and reliability in operating the voltage of microgrids have become necessary due to local sources of reactive power. Droop control technology has been successfully applied to this problem and remains popular today. This study proposes a control strategy that can be utilized to power-sharing and adjust the voltage and frequency appropriately according to the load condition. The main aim of the research is to control the frequency and voltage of microgrids under various conditions by using two algorithms, the Gray Wolf Optimizer (GWO) and Kepler Optimization Algorithm (KOA) to optimize the droop control and optimize the PI controller parameters. Simulation findings using Simulink in MATLAB demonstrate the performance of the suggested microgrid stability techniques. Finally, to evaluate the efficiency of the suggested control strategy, its results are compared with conventional methods.
Authors - Ahmed Abu-Khadrah, Munirah Ali ALMutairi, Mohammad R. Hassan, Ali Mohd Ali Abstract - The Internet of Things (IoT) devices are employed in various industries, including health care, smart homes, smart grids, and smart cities. Researchers address the intricate connection between the growth of the Internet of Things and the hazards to its security. The vast and varied features of the Internet of Things make traditional security solutions ineffective. A new model is developed to enhance IoT malware detection by combining three machine learning algorithms: KNN, Bagging, and support vector machines. The proposed model is evaluated by measuring accuracy, precision, recall and F1-score. In addition, two comprehensive datasets are utilized to evaluate the proposed model dataset. The study explores the potential of three ensemble classification models for Malware Detection. This study investigated the efficacy of a novel ensemble machine-learning approach for detecting malware within the Internet of Things (IoT) domain. The result of this research is that the accuracy on the validation set is 95.76%, the precision on the validation set is 97.01%, the recall is 94.55%, and the F1 score is 95.77%. The findings of this study indicate that the proposed model, a synergistic combination of K-Nearest Neighbours (KNN), Bagging, and Support Vector Machines (SVM), achieved a commendable overall accuracy of 95.76% in correctly classifying both malware and benign programs within the utilized IoT dataset.
Authors - Hoang Minh Tuan, Ngo Gia Quoc, Nguyen Huu Tien, Vu Thu Diep Abstract - This paper provides a method for automatically detecting fake beef by image analysis. High-quality classification models could have a major impact on ensuring food quality, supporting supply chain management in the meat industry, and preventing fraudulent commercial practices. Because low-quality meat is cheaper and more widely available than beef, it is common to use them as a substitute for fake beef. The problem is due to the differences in meat appearance, texture, mutilation, and color of cuts, as well as similarities between real beef and fake beef. These characteristics require a robust method to distinguish subtle characteristics to obtain reliable results. This paper combines the strength of Convolutional Neural Networks to detect a true classification of beef and fake beef. This model targets mobile applications and is suitable for the practical deployment of various environments.
Authors - George Kwamina Aggrey, Amevi Acakpovi, Emmanuel Peters Abstract - ERP systems are integrated information systems (IIS) popularly used among tertiary institutions in the globe. ERP has attained familiarization in certain parts of the globe due to its huge acquisition in tertiary institutions. Notwithstanding the rising acquisition, choice and execution of ERPs in higher education, there remains a scarcity in literature about their performance especially in the developing world. It is, therefore, important to further examine whether these ERPs fulfill their anticipated benefits. This paper aims to evaluate the effectiveness of KNUST's enterprise system (comprising ARMIS, Panacea, and Synergy Systems) in HEIs through a system integrative framework. A combined-method research approach was employed, collecting data from a sample of 60 respondents for both quantitative and qualitative investigation. The data were examined through partial_least_squares structural_equation_modeling (PLSSEM) and inductive_thematic_analysis. The study's results revealed that the customer/stakeholder-perspective, learning-growth-perspective, financial-perspective and system-quality-perspective significantly influence and positively relate to the effectiveness of KNUST's enterprise system evaluation in Ghanaian higher education. Internal business process, according to the findings, was the only perspective that had no significant impact on the performance of KNUST enterprise system in the Ghanaian higher education. Works on ERPs assessments, readiness, and implementations are scarce in developing world, particularly in the Ghanaian context. This study has successfully assessed the KNUST enterprise system, demonstrating its effectiveness through the research model deployed.
Authors - Jimmy Katambo, Gloria Iyawa, Lars Ribbe, Victor Kongo Abstract - The vulnerability of Southern Africa to climate variability, especially drought, places substantial pressure on agriculture, water systems, and the economy. This study explores how El Niño-Southern Oscillation (ENSO)-related Sea Surface Temperature (SST) variations influence drought patterns across the region using machine learning methods. Two approaches were taken: (i) a feature ranking of SST in comparison to twelve other climate variables and (ii) drought model performance comparisons with and without SST data. Results reveal SST’s significant and consistent impact across all climate zones, with both methods indicating that SST data, particularly in connection with ENSO phases, strongly influences drought variability, despite slight variations in its order of effect with respect to climatic zonal divisions. This underscores the value of incorporating SST in climate models for enhanced drought prediction and adaptation planning. Although limited by a focus on SST and not fully accounting for interactions with other climate factors, this research provides a solid foundation for understanding regional climate dynamics. Adding more climate indicators and studying SST’s interactions with land-based factors could help future studies make drought predictions more reliable and better prepare vulnerable areas.
Authors - Ping Luo, Kyle Gauthier, Bo Huang, Wenjun Lin Abstract - Data visualization is a critical tool for interpreting complex information, yet it often remains inaccessible to those without extensive technical and analytical skills. This study introduces a novel multi-agent system leveraging a large language model (LLM) to democratize the process of creating high-quality visualizations. By automating the stages of planning, coding, and interpretation, the system empowers users with diverse backgrounds to generate accurate and meaningful visual representations of data. Our approach employs multiple specialized agents, each focusing on different aspects of the visualization workflow, thereby enhancing the overall quality through collaborative problem-solving and contextual communication. The iterative refinement phase ensures that the visualizations meet the initial objectives and data characteristics, thus improving accuracy and relevance. This study’s modular design allows for scalability and adaptability to various data types and visualization needs, ensuring the system remains current with emerging tools and frameworks. By lowering the barriers to effective data visualization, our system supports broader data-driven decision-making across various domains, fostering more inclusive and impactful data analysis practices. Validation on two public datasets demonstrates that our multi-agent framework generates visualizations that achieve comparable or superior quality metrics when benchmarked against human expert analysis.
Authors - Nelly Galora-Moya, Paola Ramos-Medina, Elsa Hernandez-Cherrez, Javier Sánchez-Guerrero Abstract - This study analyzes the necessity and impact of English for Specific Purposes (ESP) courses on B1+ level students, integrating gamification as an innovative approach through the use of the Duolingo platform. Surveys were ad-ministered to students and faculty members from various departments to gather their perceptions on the relevance and feasibility of gamified ESP courses. Additionally, a preliminary diagnostic test was conducted to assess technical vocabulary knowledge in a gamified environment. The results show a general consensus on the importance of ESP courses, highlighting Duolingo as an effective tool for personalizing learning and enhancing linguistic competencies in specific contexts. Significant gaps in students' linguistic skills were also identified, justifying the incorporation of this methodology. The paper proposes a collaborative program between the faculties of the Technical University of Ambato (UTA) and the Language Center, with Duolingo playing a central role in designing a gamified curriculum to bridge the gap between academic English and the specialized skills required in professional settings. This approach not only improves academic performance but also equips graduates with the linguistic skills necessary to compete in the global job market. Gamification, by combining motivation and interactivity, fosters autonomous and collaborative learning. This work contributes to the current debate on the use of technology in ESP teaching, emphasizing gamification as a key strategy for personalizing learning. Future research should assess the longitudinal impact of gamified platforms like Duo-lingo on academic performance and graduates’ career trajectories.
Authors - Reem M. Zemam, Nahla A. Belal, Aliaa Youssif Abstract - According to the World Health Organization’s statistical data for 2024, breast cancer is the most often diagnosed cancer among women.Between 2020 and 2024, approximately 37,030 new instances of invasive breast cancer were documented in women.Recent advancements in deep learning have shown considerable potential to improve the accuracy of breast cancer diagnosis, ultimately aiding radiologists and clinicians in making more precise decisions.This study presents a strategy that creates a highly dependable ultrasound analysis reading system by comparing the powerful processing capabilties of CNNs with 4 pretrained models (Transfer Leraners). The models employed were the DenseNet 169, ResNet 152, MobileNet V2, and Xception. To assess the effectiveness of the proposed framework, experiments were conducted using established bench- mark datasets (BUSI datasets). The suggested framework has demonstrated superior performance compared to previous deep learning architectures in precisely identifying and categorizing breast cancers in ultrasound images. Upon comparison of the specified deep learning models, DenseNet 169 had the maximum performance with an accuracy of 99.7%. This surpasses the research undertaken in the literature. This research employs advanced deep learning algorithms to enhance breast cancer diagnostic outcomes, decreasing diagnosis time and facilitating early treatment.
Authors - Eduardo Pinos-Velez, Adriana Martinez-Munoz, Dennys Baez-Sanchez Abstract - Prolonged sitting is a significant contributor to various health issues, including pressure ulcers, lower back pain, and circulatory disorders. This paper provides an analysis of these pathologies, examining their underlying causes, physiological impacts, and the compounding role of risk factors such as physical inactivity and poor posture. Furthermore, the study evaluates technological solutions designed to mitigate these risks. These include advanced sensor-integrated cushions and alternating pressure systems that facilitate weight redistribution to prevent tissue damage.
Authors - Desmond Kwadjo Kumi, Sheena Lovia Boateng Abstract - This study uses the utility theory framework to investigate cryptocurrency awareness and usage behavior among Gen Z in Ghana. Data were collected from 700 individuals in the educational sector of Ghana, aged 18-25 years, through purposive and snowball sampling, using structured questionnaires, with 657 usable responses. Data was analyzed using SPSS, AMOS, and Hayes Process Macro. The results revealed that perceived benefits significantly indirectly affect cryptocurrency usage behaviour via cryptocurrency’s perceived value. Perceived risks lessen the influence of perceived benefits on perceived value, whereas personal innovativeness improves this link. The survey further revealed a very high awareness of Bitcoin and other cryptocurrencies, but comparatively lower awareness of the entire blockchain technology. Whereas, awareness, attitudes toward and ownership of cryptocurrencies were higher among males than females, thus showing a gender gap in the awareness and ownership of digital assets. This study is arguably one of the few sources with insights into applying the utility theory to understand cryptocurrency usage behaviour among Gen Z in a developing economy like Ghana. Practitioners and policymakers could therefore, tailor strategies that address awareness and ownership gaps and optimize utility dimensions.
