SACM - United States of America
Permanent URI for this collectionhttps://drepo.sdl.edu.sa/handle/20.500.14154/9668
Browse
1838 results
Search Results
Item Restricted A Facial Expression-Aware Edge AI System For Driver Safety Monitoring(Saudi Digital Library, 2025) Almodhwahi, Maram; Wang, BinThis dissertation presents a driver monitoring system (DMS) that integrates emotion recognition to address critical issues in road safety. Road safety has become a global concern due to the significant increase in vehicle numbers and the rapid growth of transportation infrastructure. The number one cause of road accidents is human error, with a 90% ratio, with common contributing factors like distraction, drowsiness, panic, and fatigue. Traditional DMS approaches often fall short in identifying these emotional and cognitive states, limiting their effectiveness in accident prevention. To address these limitations, this research proposes a robust, deep-learning-based DMS framework designed to identify and respond to driver emotions and behaviors that may compromise safety. The proposed system utilizes advanced convolutional neural networks (CNN), specifically the inception module and Caffe-based ResNet-10 with a single-shot detector (SSD), to perform efficient facial detection and classification. These chosen model structures helped balance computational efficiency and accuracy. The DMS is trained on an extensive, diverse dataset comprising approximately 198,000 images and 1,600 videos sourced from multiple public and private datasets, ensuring the system’s robustness across a range of emotions and real-world driving scenarios. Emotions of interest include high-risk states such as drowsiness, distraction, and fear, alongside neutral conditions, and the model can perform well in different conditions, including low-light and foggy/blurry environments. Methodologically, the system incorporates essential data preprocessing techniques such as resizing, brightness normalization, pixel scaling, and noise reduction to optimize the model’s performance. On top of that, data augmentation and grayscale conversion improves the dataset’s variability, allowing the decrease of computational costs without sacrificing accuracy. This approach enabled the model to achieve high performance metrics, with an overall accuracy of 98.6% , an F1-score of 0.979, precision of 0.980, and recall of 0.979 across the four primary emotional states. This research contributes to the field by offering a less invasive, real-time solution for monitoring high-risk driver behaviors and providing insights for further advancements in automated driver assistance technologies. Future directions include optimizing the system for microcontrollers with low power consumption and implementing alerts for high-risk states to further mitigate accident risks, as well as including a multi-modal fusion of data from different sources (Infrared Camera, and a Microphone) to increase emotion recognition accuracy, which leads to taking better control and initiating more efficient proactive interventions.6 0Item Restricted A HETEROGENEOUS INTEGRATED PHOTONIC PLATFORM FOR HIGH SENSITIVITY SENSORS(Georgia Institute of Technology, 2025) Alsaggaf, Abeer; Adibi, AliHorizontal slot microdisk resonators provide a new pathway for integrated photonic sensing by combining ultra-high optical performance with strong interaction between light and analytes. In this work, a complete design and nanofabrication process was developed to realize these devices, which were validated through chemical and biomolecular sensing experiments. The approach demonstrated reliable detection of both refractive index changes and surface-adsorbed biomolecular layers, including the clinically relevant biomarker Troponin I. These results highlight the horizontal slot architecture as a powerful platform for lab-on-chip technologies with applications in healthcare diagnostics, environmental monitoring, and portable point-of-care systems.8 0Item Restricted Toward an Integrative Study of Human-AI Interaction(Saudi Digital Library, 2025) الصبي, محمد عبدالرحمن; Almaatouq, AbdullahAs artificial intelligence (AI) systems are increasingly embedded in the workflows of individuals and groups, designers and researchers of human-AI interaction (HAI) navigate a vast design space of possible configurations, making decisions that span algorithmic parameters, interface choice, and interaction protocols. This thesis develops an integrative approach that examines how design factors combine and interact to determine the outcomes of human-AI collaboration. Chapter 1 synthesizes prior HAI research into a coherent design space framework encompassing algorithms, interfaces, users, and task settings, motivating a research program for systematic exploration of interdependencies between these factors. Chapters 2 and 3 turn to group-AI interaction through large-scale behavioral experiments. Chapter 2 investigates how social information---both direct conversation and peer behavior indicators---affects individual reliance on algorithmic decision support. The study reveals that while social information modulates the effects of performance feedback and model explanations on reliance, it does not improve predictive accuracy, illuminating critical tensions between social mechanisms and system design. Chapter 3 examines large language models as facilitators of group deliberation in hidden profile tasks. While LLM facilitation increased information sharing volume, density, and breadth, it did not improve decision quality, highlighting fundamental challenges in group-AI system design beyond information aggregation. Chapter 4 advances an integrative approach to HAI research, emphasizing shared design spaces, systematic exploration strategies, and predictive models that generalize across contexts. The chapter provides methodological guidance and a tractable roadmap for advancing this integrative research agenda, laying the foundation for a more context-aware science of human-AI collaboration.16 0Item Restricted Modeling and Experimental Validation of Time Delay Effects in Nonlinear Digital Feedback-based Resonators(Saudi Digital Library, 2025) Koshak, Amro; Bajaj, NikhilThis work demonstrates the use of nonlinear digital feedback on a microelectromechanical system (MEMS) resonator to create dynamical bifurcation behavior, and through modeling and experiment, considers the significant effect of inherent and controlled feedback delays on the resulting dynamical system. The resonator used to experimentally validate the theoretical predictions is a piezoelectric microcantilever fabricated through the PiezoMUMPS process. Two equations of motion were studied and solved using two different perturbation methods. The first method was the Krylov-Bogolyubov (KB) averaging method, while the second equation of motion used the method of multiple time scales. The KB method focused on studying the effects of a delayed cubic stiffness feedback on system dynamics. The multiple time scales approach was also considered, and applied to a linear delayed feedback as an addition to the delayed cubic stiffness in the feedback. The experimental results were attained through the use of the Moku:Pro's Field programmable gate array (FPGA), which digitally implemented a tunable feedback to the resonator. Frequency sweeps were conducted under varying feedback parameters that demonstrated strong agreement between the analytical and experimental results. The results validated the flexibility of employing the FPGA and the effects of varying the feedback parameters digitally, and point towards a flexible means of implementing nonlinear sensing schemes with resonators in the future.17 0Item Restricted Optimizing Veterinary Drug Residue Monitoring in U.S. Cattle through Trend Analysis and Risk-Based Frameworks(Saudi Digital Library, 2025) AlWahaimed, Abdullah Saud; Eifert, JosephVeterinary drug residues in U.S. cattle continue to raise important food safety, regulatory, and public health concerns. This dissertation integrates three complementary investigations to examine residue trends, evaluate human health risks, and propose an improved framework for national monitoring. First, national sampling data from the USDA Food Safety and Inspection Service (FSIS) National Residue Program (NRP) for 2021–2023 were analyzed to identify frequently detected residues across cattle types. Cattle accounted for 92% of all positive samples across species sampled (sheep and goats, pigs, poultry, and fish), totaling 3,107 out of 3,391 detections. Dairy cows had the highest number of detections within cattle (1,264 out of 3,107). Desfuroylceftiofur, penicillin, flunixin, and several sulfonamides were among the most commonly detected violative residues. Second, a multifactorial risk-ranking model was developed to assess the potential public health risks associated with fifteen high-priority veterinary drugs detected in cattle. The model integrated hazard evidence, exposure frequency, and severity of adverse outcomes across eleven risk domains. Sulfonamides consistently ranked as the highest-risk group, while β-lactams and non-steroidal anti-inflammatory drugs (NSAIDs) ranked as intermediate-risk due to frequent detection despite moderate inherent toxicity. Finally, an adaptive decision support framework was created to improve veterinary drug residue monitoring by incorporating drug-specific, species-specific, and exposure-related determinants into a unified evaluation system. The framework provides clear thresholds for prioritizing high-risk drug species combinations and can support more efficient risk-based sampling strategies aligned with the Food and Agriculture Organization (FAO) and the World Health Organization (WHO), as well as the Codex Alimentarius Commission. Collectively, these studies demonstrate the need for continuous monitoring of frequently detected residues, improved prioritization of high-risk analytes, and an adaptive, evidence-driven approach to sampling. The findings offer actionable recommendations to strengthen residue-control programs, enhance regulatory responsiveness, and protect public health.4 0Item Restricted Distance Based Statistical Methods and Outlier Detection in Large-Scale Electrophysiology Data(Saudi Digital Library, 2025) Asiri, Zahra; Maia, PedroLarge-scale electrophysiology experiments produce high-dimensional local field po- tential (LFP) datasets whose size and heterogeneity challenge classical analysis methods. This thesis develops a unified and scalable computational framework for comparing mul- tichannel rodent LFP recordings collected under formalin injection and electrical stimu- lation. We begin by outlining the biological context and formalizing the core research questions in precise mathematical terms. Building on this foundation, we introduce three complementary methodological contributions. First, a window-based fusion framework enables scalable column-wise comparison of large matrices by replacing quadratic-distance computations with segmented, statistically fused evidence. Second, a landmark-based clustering approach provides efficient approximations to pairwise Euclidean distances, with explicit operation-count models and a practical scaling rule that generalizes across synthetic and real data. Third, a row-wise analysis framework based on Elastic-Net PCA and CCA yields low-variance geometric embeddings that support reliable statistical comparison between baseline and post-treatment recordings. To detect perturbation-induced anomalies, we develop the Combined Outlier Score (COS), an ensemble of nine unsupervised detectors that integrates geometric, probabilis- tic, and density-based signals into a unified anomaly measure. Applied to the rodent migraine recovery dataset, the full framework identifies rest intervals that statistically match baseline structure, quantifies deviations following stim- ulation, and reveals interpretable temporal recovery patterns. The results demonstrate that segmentation-based fusion, landmark approximation, row-wise embeddings, and en- semble outlier detection together form a robust and computationally efficient toolkit for analyzing high-dimensional neural data. This thesis provides a coherent methodological foundation for scalable similarity assessment and anomaly detection in large electrophysiology datasets, with applicability to a broad range of big-data time-series domains.12 0Item Restricted Transcription-Coupled Removal of Formaldehyde-Induced DNA-Protein Crosslinks(Saudi Digital Library, 2025) Alshareef, Duha; Campbell, ColinDNA-protein crosslinks (DPCs) form following exposure to various alkylating agents including environmental carcinogens, cancer chemotherapeutics, and reactive aldehydes. If not repaired, DPCs can interfere with key biological processes such as transcription and replication and activate programmed cell death. A growing body of evidence implicates nucleotide excision repair (NER), homologous recombination, and other mechanisms in the removal of DPCs. However, the effects of genomic context on DPC formation and removal have not been comprehensively addressed. Using a combination of next generation sequencing and DPC enrichment via protein precipitation, I showed that, unlike spontaneous DPCs, formaldehyde-induced DPCs are non-randomly distributed across the human genome, based on chromatin state. I also showed that the efficiency of DPC removal correlates with transcription at loci transcribed by RNA polymerase II. Using repair mutant cell lines, I found that efficient removal of chromosomal DPCs requires both the Cockayne syndrome group B gene as well as ‘downstream’ transcription-coupled-NER factor xeroderma pigmentosum group A gene. In contrast, I found that loci transcribed by RNA polymerase I showed no evidence of transcription-coupled DPC removal. Finally, using pharmacological inhibition of Rad5, I was able to show a reduced efficiency of DPC removal. Taken together, the results indicate that complex interactions between chromatin organization, transcriptional activity, and numerous DNA repair pathways dictate genomic patterns of DPC formation and removal.14 0Item Restricted COST-EFFECTIVENESS ANALYSIS OF NOVEL ADD-ON CARDIORENAL PROTECTIVE THERAPIES IN PATIENTS WITH TYPE 2 DIABETES AND CHRONIC KIDNEY DISEASE WITH ALBUMINURIA(Saudi Digital Library, 2025) Alharbi, Haifa; Sullivan, PatrickType 2 diabetes mellitus (T2DM) is the leading cause of chronic kidney disease (CKD) in the United States. CKD increases the risk of cardiovascular (CV) events, kidney failure, and death. Standard of care (SoC), primarily renin–angiotensin system inhibitors, slows but does not prevent progression. Evidence recently shows cardiorenal benefits of sodium–glucose cotransporter-2 inhibitors (SGLT2i), glucagon-like peptide-1 receptor agonists (GLP-1 RAs), and nonsteroidal mineralocorticoid receptor antagonists (ns- MRAs) as add-on strategies; yet, their cost-effectiveness remains a concern. This study evaluated the cost-effectiveness of five treatment strategies in adults with T2DM and CKD with albuminuria in the U.S. from a healthcare payer perspective. The strategies included SoC alone, SoC + SGLT2i, SoC + SGLT2i + ns-MRA, SoC + SGLT2i + GLP-1 RA, and SoC with all three therapies combined. A Markov model projected CKD progression using KDIGO heat map classification risks, CV events, kidney failure requiring replacement therapy (KFRT), and mortality over a lifetime horizon. Costs, quality-adjusted life years (QALYs), incremental cost-effectiveness ratios (ICERs), and net monetary benefit (NMB) were estimated using U.S. data. In the base case, SoC yielded the lowest lifetime cost and QALYs ($283,833, 9.31 QALYs). Adding an SGLT2i improved outcomes to 10.36 QALYs at $312,284, with an ICER of $27,095/QALY, below the $100,000 willingness-to-pay (WTP) threshold. Adding an ns- MRA increased QALYs to 10.93 with an ICER of $160,479/QALY. Strategies including GLP-1 RAs produced ICERs ranging from $230,000 to $1.3 million. NMB analysis identified SoC + SGLT2i as the most cost-effective strategy ($723,745 at $100,000/QALY). Probabilistic analysis showed SoC + SGLT2i was cost-effective in all simulations, while SoC + SGLT2i + ns-MRA had an 82% probability of being costeffective at $100,000/QALY. A 50% SGLT2i price reduction resulted in the SoC + SGLT2i strategy becoming dominant and improved the cost-effectiveness of adding ns- MRAs. Threshold analyses indicated that ns-MRA prices would need to decrease by 6– 60% and GLP-1 RA prices by 33–72% to align with WTP benchmarks. Overall, SGLT2i is the most cost-effective strategy, ns-MRAs could become cost-effective if prices are lowered, and GLP-1 RAs would require substantial discounts.1 0Item Restricted END-TO-END DIRECT CURRENT FOR STANDALONE POWER NETWORK(Saudi Digital Library, 2025) Aldarsi, Eyad; Singh, RajendraOriginally, the electrical energy used to be generated in form of direct current (DC) system, which was used for the first time by Thomas Edison in 1882. However, due to the growth of the loads and the increase of transmission distances, the generation of electric energy in form of alternating current (AC) system has become the standard in the entire world for the last century. The emergence and constant technological advancements of power electronics during the last century coupled with the photovoltaics (PV) commercialization, which generates DC power, as well as grid-scale battery energy storage system, which stores DC power, have demonstrated the low-cost sustainability of electricity generated in DC power form. In this document, the author has introduced the stand-alone power network for the emerging concept of community solar that is based on the end-to-end DC power network as a potential solution for phasing away gradually from AC power grid infrastructure; and bring back the concept of centrally located power generation based on the original idea of Thomas Edison. This off-grid power network consists of solar PV farm coupled with battery-based energy system, which generates and stores DC power respectively as mentioned earlier, and can supply to the utilization site with electricity all year long. The performance of this independent microgrid shows no deficiency in the supplied electricity in the first year for the project as well as after thirty years. The results of this dissertation show that the usage of DC power network in the form of end-to-end has a major advantage as compared to the conventional grid in term of energy efficiency as well as energy surplus which results directly in financial resource preserving.17 0Item Restricted Saudi Early Childhood Teacher Educators’ and Preservice Teachers’ Perspectives of Early Field Experiences(Saudi Digital Library, 2025) Alodwani, Amani; Sophia, HanThis qualitative descriptive study explores how teacher educators perceive the integration of field experiences within the curriculum and their perspectives on the preservice early childhood education (ECE) teachers they supervise. It also examines preservice teachers’ views regarding their field experiences and how these experiences contribute to their professional preparation. The Ministry of Education (MOE) emphasizes that teacher preparation programs should provide extended field experiences that connect theoretical coursework with practical application. Semi-structured interviews were conducted with three Saudi teacher educators and four preservice teachers from Kingdom University. Thematic analysis using inductive coding was employed to identify recurring themes and differences between the two groups. Findings revealed that both groups recognized the importance of field experiences but differed in how they interpreted and applied them. Teacher educators highlighted challenges related to supervision and assessment, while preservice emphasized time conflicts and the value of hands-on learning. The results underscore the need for culturally grounded and collaborative supervision practices that effectively bridge theory and practice. These insights contribute to enhancing the design and implementation of field experiences in Saudi early childhood education, thereby promoting professional growth and culturally responsive teaching.13 0