Mayda Alrige

The Kingdom of Saudi Arabia launches the Saudi Architecture Map, which includes 19 architectural styles for the country’s main regions. These architectural styles are shaped by distinct climatic, material, and cultural determinants across the country. A deep dive into the architectural elements that define these styles reveals many similarities and nuances, making manual classification of modern buildings challenging. Despite the kingdom’s rich heritage, and since the launch of the Saudi Architecture Map and its contained styles and corresponding design guidelines, no operational digital pipeline currently exists for automated, quantitative, and reproducible classification of façade imagery against these styles. This limits heritage documentation and compliance monitoring at scale. This study attempts to fill this gap. We developed a deep-learning classification framework for building façade imagery considering the Saudi architectural styles. We curated a dataset of 3750 façade images, comprising 2320 full-building façade images and 1430 architectural-element images. We collected the data from online heritage repositories. We benchmarked five convolutional backbones (ResNet50, EfficientNetB0, EfficientNetV2M, ConvNeXtBase, and InceptionV3) under a unified two-stage transfer learning strategy. The first stage trained classifier heads on architectural-element images with frozen backbone weights, and the second stage fine-tuned the top 30% of backbone layers on full-building façade images. Among the five CNN backbones, ConvNeXtBase achieved the strongest performance across all four metrics, achieving 90.75% in overall accuracy, 88.06% in F1-Score, 92.05% in producer accuracy, and 86.65% in user accuracy. Ten of the eighteen styles exceeded 90% F1, and two styles—Bisha Desert and Tabuk Coast, both with very small corrected test counts (n = 6 and n = 1, respectively)—fell below 70%. These results indicate that ConvNeXtBase with two-stage transfer learning is a promising step toward automated documentation of Saudi architectural styles. Yet, further validation through ablation testing or calibration is needed before deployment for compliance monitoring.

Individual actions play a pivotal role in climate change, one of the most urgent global challenges, as daily behaviors generate substantial greenhouse gas emissions. Saudi Arabia, in particular, demonstrates its strong commitment to environmental sustainability through the Saudi Green Initiative and Middle East Green Initiative, aiming for net-zero emissions by 2060 and advancing reforestation, land conservation, and renewable energy under Vision 2030. However, many Saudi individuals remain unaware of the environmental consequences of their choices, including transportation, energy consumption, and lifestyle habits. To address this gap, this study developed GreenKSA, the first Arabic-supported gamified mobile application designed to promote pro-environmental behavior within the Saudi culture. The app integrates gamification elements grounded in Self-Determination Theory (SDT) and Trans-Theoretical Model (TTM) in an attempt to bridge the gap between theory and practice. GreenKSA delivers multimedia content—short videos and infographics—demonstrating sustainable routines in households, workplaces, and mobility. The design and user experience of GreenKSA were evaluated in a pilot study of 10 participants. The results indicated high usability (SUS = 91.25) and a positive overall user experience. By combining theory-driven design with culturally relevant gamification elements, this study contributes to digital sustainability interventions and aligns with the global Sustainable Development Goals SDG 12: Responsible Consumption and Production, and SDG 13: Climate Action.

Large language models have revolutionized various sectors, including education and healthcare, by demonstrating significant advancements in producing human-like text and enhancing translation accuracy. Despite these advancements, the conventional process of cross-cultural instrument translation remains labor-intensive, expensive, and heavily dependent on the expertise of human professionals, including translators and healthcare specialists. The objective of this study is to address the translation challenges in health-related instruments. This study introduces a Human and Large Language Models Collaboration Framework, called H-LLMCF, designed to automate the key steps of instrument translation while maintaining linguistic precision and cultural appropriateness. The proposed framework integrates human expertise and LLM capabilities across two phases: AI-guided translation and human-guided evaluation and refinement. When applied to the translation of two validated instruments in the health sector from English to Arabic, the framework demonstrated promising outcomes. The proposed framework achieved BLEU scores of ~0.54 and ~66 for the two instruments. SBERT-based cosine similarity exceeded 0.85 for 80% of items, with a mean of ~0.92 and ~0.96, indicating strong semantic alignment with human translations. Human evaluation of the 35-item HPV Knowledge Assessment instrument yielded an average I-CVI of 1.00, with 91.42% of items rated as fully acceptable by both reviewers. For the KIDMED instrument, out of 16 items, 15 (93.75%) received a rating of 3 from both reviewers, resulting in an I-CVI of 1.00 for those items. Most discrepancies were linked to translation and fluency adaptation. This highlights the potential of LLMs to enhance cross-cultural translation processes while maintaining translation quality.

Ensuring hygiene compliance in regulated environments—such as food processing facilities, hospitals, and public indoor spaces—requires reliable detection of personal protective equipment (PPE) usage, including gloves, face masks, and hairnets. Manual inspection is labor-intensive and unsuitable for continuous, real-time enforcement. This study benchmarks three lightweight object detection models—YOLOv8n, YOLOv10n, and YOLOv12n—for automated PPE compliance monitoring using a large curated dataset of over 31,000 annotated images. The dataset spans seven classes representing both compliant and non-compliant conditions: glove, no_glove, mask, no_mask, incorrect_mask, hairnet, and no_hairnet. All evaluations were conducted using both detection accuracy metrics (mAP@50, mAP@50–95, precision, recall) and deployment-relevant efficiency metrics (inference speed, model size, GFLOPs). Among the three models, YOLOv10n achieved the highest mAP@50 (85.7%) while maintaining competitive efficiency, indicating strong suitability for resource-constrained IoT-integrated deployments. YOLOv8n provided the highest localization accuracy at stricter thresholds (mAP@50–95), while YOLOv12n favored ultra-lightweight operation at the cost of reduced accuracy. The results provide practical guidance for selecting nano-scale detection models in real-time hygiene compliance systems and contribute a reproducible, deployment-aware evaluation framework for computer vision in hygiene-critical settings.

Maple syrup urine disease (MSUD) is a metabolic disorder characterized by a difficulty to digest and process proteins necessary for growth. To monitor and maintain the ideal growth of children with MSUD, caregivers need to carefully control the consumption of harmful branched-chain amino acids (BCAAs). The dietary limits of amino acids for MSUD patients are recommended and controlled by pediatricians and metabolic dietitians according to age, height, weight, and the prevailing percentage of amino acids in the body. This study introduces an intelligent dietary tool called MSUD Baby Buddy for caregivers of MSUD patients that tracks the amino acids intake out of baby formulas for babies 0–6 months old. This tool aims to provide accurate recommendations of the appropriate daily intake of protein and BCAAs based on the patients’ data, plasma BCAAs, and formula preferences. We use a knowledge-based system, including knowledge acquisition and verification, as well as knowledge management tool validation, and the ripple-down rules are employed for building the system. MSUD Baby Buddy can support the maintenance of adequate amino acid levels and increase awareness about the control of BCAAs. The average usability of MSUD Baby Buddy is 84.25, indicating that the tool is intuitive and may help caregivers to easily determine the recommended doses of formula based on patients’ biometric data and preferred formula. On the other hand, interviews with metabolic dietitians revealed some drawbacks, which were addressed to further improve the tool. MSUD Baby Buddy is expected to help caregivers of MSUD patients to independently track nutrient intake and reduce the number of visits to the pediatrician and metabolic dietitian.

Background and Objectives: In this study, we assessed the potential impact of employee empowerment on health care workers' performance during the novel coronavirus SARS-CoV-2 (COVID-19) pandemic. In particular, we aimed to determine the empowerment practices that would have the greatest positive effect on employee performance. Understanding the relationship between performance and empowerment can help health care providers better manage worker stress during any global crisis. This understanding is crucial in guiding policies and interventions aimed at maintaining health care workers' psychological well-being and their overall performance. Methods: This cross-sectional study evaluated the relationship between employee empowerment and performance, determining the best empowerment practices for health care leaders to utilize. Frontline health care workers (n = 100) selected using convenience and snowball sampling completed the survey between March 15 and 31, 2020. This is the period when the pandemic just started to accelerate in Saudi Arabia. We conducted Pearson's correlation analysis to assess whether there was a relationship between performance and health care workers' empowerment practice, and stepwise linear regression analysis to investigate the impact each of these empowerment practices on health care workers' performance. Results: Our results indicate that health care workers' performance can be expected to increase the most through 2 empowerment practices: giving employees the discretion to change work processes and offering performance-based rewards (R ² = 0.301, P < .05). Conclusion: Our findings suggest that health care leaders must invest in these 2 practices to better equip frontline health care workers. During a global crisis, additional discretion granted to employees helps reduce their anxiety and burnout and hence empowers them with the flexibility to adapt to unforeseen circumstances and improve the quality of their interactions with health service recipients.

For years, several countries have been concerned about how to dispose of unused pharmaceuticals that can endanger human health and the environment. Moreover, some people are in desperate need of medical attention and medications, but they lack the financial resources to obtain them. In Saudi Arabia, there are no take-back medicine programs, and there is no published research on how medications properly are disposed. The aim of this research is to use the power of artificial intelligence to assist in the proper management and disposal of expired and unused medications and to develop a prototype device for collecting medication by automatically classifying medications for proper disposal and donation. In this research, artificial intelligence technologies such as web-based expert systems, image recognition and classification algorithms, chatbots, and the internet of things are used to assist in a take-back medications program. In conclusion, the prototype design of a web-based expert system and the device reduced improper disposal risks by providing significant advice on the safe disposal of unwanted pharmaceuticals. By using an organized method of collecting expired medications, the benefits were made possible.

BACKGROUND Oral diseases have been described by the World Health Organization (WHO) as the most prevalent diseases globally, affecting some 3.5 billion people. This leads to significant health and economic burdens and can impact the quality of life of affected individuals. Therefore, dentists have a great responsibility to efficiently diagnose and determine the best treatment option. However, some do not have the experience and knowledge to make the right clinical decisions. For this reason, artificial intelligence (AI) techniques, mainly rule-based systems, have been used in dentistry to aid physicians in making faster and more reliable decisions. OBJECTIVE This scoping review aims to explore and summarize the application of rule-based systems widely employed in dentistry and to evaluate their performance and practical significance. METHODS We conducted a scoping review following the methodology of Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) on five databases: Web of Science, Scopus, Google Scholar, Saudi Digital Library, and the IEEE Xplore. We searched for literature published in English up to October 2021. Two reviewers evaluated each potentially relevant study for inclusion/exclusion criteria, and any discrepancies were resolved by a third researcher. RESULTS Of 303 studies, 19 fulfilled this review’s inclusion criteria. We identified two domains based on the methodology used in the included studies: (i) uncertainty management approaches employed in the rule-based system (n = 16) and (ii) integrating machine learning techniques with the rule-based system (n = 5). The vast majority of included publications used fuzzy logic to manage uncertainty (n = 11). A hybrid fuzzy rule-based system and neural network achieved the highest accuracy of 96%. From a medical perspective, the articles were aimed at diagnosis (n = 11), treatment (n = 3), and both diagnosis and treatment (n = 4), while less attention was paid to detection and classification (n = 1). The review also found that periodontology was the most commonly addressed specialty. CONCLUSIONS In an analysis of the current literature, rule-based systems were found reliable to assist dental practitioners in decision-making. Clinical decision-making involves a high level of uncertainty, which explains the tendency to use fuzzy logic in rule-based systems. These systems can also be used as educational tools primarily for both dental interns and less experienced general dentists to aid in making reliable decisions.