Explainable Transformer-Based Framework for Glaucoma Detection from Fundus Images Using Multi-Backbone Segmentation and vCDR-Based Classification Hind Alasmari, Ghada Amoudi, Hanan Alghamdi Diagnostics, 2025 Glaucoma is an eye disease caused by increased intraocular pressure (IOP) that affects the optic nerve head (ONH), leading to vision problems and irreversible blindness. Background/Objectives: Glaucoma is the second leading cause of blindness worldwide, and the number of people affected is increasing each year, with the number expected to reach 111.8 million by 2040. This escalating trend is alarming due to the lack of ophthalmology specialists relative to the population. This study proposes an explainable end-to-end pipeline for automated glaucoma diagnosis from fundus images. It also evaluates the performance of Vision Transformers (ViTs) relative to traditional CNN-based models. Methods: The proposed system uses three datasets: REFUGE, ORIGA, and G1020. It begins with YOLOv11 for object detection of the optic disc. Then, the optic disc (OD) and optic cup (OC) are segmented using U-Net with ResNet50, VGG16, and MobileNetV2 backbones, as well as MaskFormer with a Swin-Base backbone. Glaucoma is classified based on the vertical cup-to-disc ratio (vCDR). Results: MaskFormer outperforms all models in segmentation in all aspects, including IoU OD, IoU OC, DSC OD, and DSC OC, with scores of 88.29%, 91.09%, 93.83%, and 93.71%. For classification, it achieved accuracy and F1-scores of 84.03% and 84.56%. Conclusions: By relying on the interpretable features of the vCDR, the proposed framework enhances transparency and aligns well with the principles of explainable AI, thus offering a trustworthy solution for glaucoma screening. Our findings show that Vision Transformers offer a promising approach for achieving high segmentation performance with explainable, biomarker-driven diagnosis.
Adapting teaching and learning with existing generative AI by higher education Students: Comparative study of Zayed University and King Abdulaziz University Dina Tbaishat, Ghada Amoudi, Maha Elfadel Computers and Education Artificial Intelligence, 2025 This study examines the role of higher education students’ perceptions in adapting Generative AI (GenAI) tools for teaching and learning, with a particular focus on the factors that influence student satisfaction and engagement. A comparative approach is adopted, exploring student experiences at Zayed University (ZU) in the UAE and King Abdulaziz University (KAU) in Saudi Arabia. The principal variables of interest, including Expected Benefits (EB), University Support (US), Ethical Awareness (EA), and Technology Self-Efficacy (TSE), are examined, with particular attention to their direct and mediated influences through Behavioral Intention (BI) on student satisfaction (SS). Data were collected through surveys and analyzed using SmartPLS-4. The findings reveal notable similarities and differences between the two universities. At both ZU and KAU, BI demonstrated the strongest direct influence on SS, confirming its central role. EB and TSE significantly impacted SS both directly and indirectly through BI in both contexts, although their effects were stronger at KAU. Conversely, US and EA showed no significant direct or mediated effects on SS at either institution. R 2 values indicated substantial explanatory power of the model, and Q 2 values confirmed strong predictive relevance. These results suggest that while the core drivers (BI, EB, TSE) are consistent across contexts, institutional and cultural factors shape their relative impact. The findings highlight the importance of integrating GenAI tools into teaching practices and emphasize the role of student motivation, confidence, and institutional support in fostering effective adoption and enhancing learning experiences. Institutions should prioritize enhancing students' perceived benefits and technological self-efficacy by providing practical training and demonstrating the value of GenAI tools. These factors significantly influence student satisfaction and engagement, particularly in culturally distinct settings.
Yaqeen: A Mobile-Based Deep Learning Framework for Prayer Posture Recognition Ghada Amoudi, Ritaj Albogami, Rahma Alsharif, Bushra Alsulami Proceedings of IEEE ACS International Conference on Computer Systems and Applications Aiccsa, 2025 Prayer (Salah) is the second pillar of Islam, and Muslims are required to perform it five times a day. Given the significance of this act of worship, considerable effort was devoted to teaching how to perform prayers correctly. Despite the availability of numerous educational resources online and on social media, many new Muslims still encounter difficulties in understanding and effectively applying these instructions. In this research, we developed “Yaqeen,” a real-time mobile system for recognizing Islamic prayer postures aimed at supporting new Muslims in learning prayers independently. The system leverages a convolutional neural network (CNN) trained on a labeled dataset of prayer images. The VGG16 model was selected for deployment due to its superior accuracy and stability, achieving 93% classification accuracy. The trained model was converted into TensorFlow Lite format and integrated into a Flutter-based mobile application, enabling low-latency, on-device inference without requiring an internet connection. This work demonstrates the practical potential of AI-powered human activity recognition for educational purposes, setting a foundation for broader applications in this field.
Leveraging AI-Generated Content for Synthetic Electronic Health Record Generation With Deep Learning-Based Diagnosis Model Sayed Abdel-Khalek, Abeer D. Algarni, Ghada Amoudi, Salem Alkhalaf, Fahad Mohammed Alhomayani, Shankar Kathiresan IEEE Transactions on Consumer Electronics, 2025 Consumer electronics have transformed the way we interact with technology, improving convenience and connectivity in day-to-day lives. In the healthcare sector, recent technologies have resulted in enhanced diagnosis, treatment, and patient care. Wearables, artificial intelligence-based data analytics, and telemedicine transform the way of monitoring and managing health, fostering a proactive approach to well-being. The popularity of ChatGPT is proven great potential for AI-generated content (AIGC) that has formed a major impact on the artificial intelligence (AI) community and accelerates the reconsidering of the prospects of general AI. The AIGC is also exposed as a considerable scope to impulse healthcare electronics (HE). Although generative AI has achieved popularity like the formation of images, it could be employed for producing synthetic tabular information. The production of synthetic electronic health records (EHR) undertakes to increase the utilization of machine learning (ML) methods that commonly function with massive quantities of data. ML will identify non-intuitive classifier patterns that permit a new integration of patient feature predictive ability. Currently, deep learning (DL) techniques are effectively utilized in EHR data from medical domains. DL methods excellently captured the significant and beneficial features and patterns from the comprehensive medical information in EHR data. This study presents AI-generated content for Synthetic Electronic Health Record Generation with a Deep Learning-based Diagnosis (SEHRG-DLD) Model. The focus of the SEHRG-DLD technique is to initially generate the synthetic EHR data and then analyze the medical data for disease diagnosis using the DL model. The SEHRG-DLD technique comprises a two-stage process: synthetic data generation and disease diagnosis. At the initial stage, the SEHRG-DLD technique uses the ChatGPT tool to generate synthetic EHR data. Then, the SEHRG-DLD technique undergoes the disease diagnosis process using three sub-processes namely Harris Hawks Optimization (HHO) based feature selection, deep belief network (DBN) based classification, and Golden Jackal Optimization (GJO) based hyperparameter tuning. The application of the HHO and GJO algorithms helps in accomplishing enhanced diagnostic performance of the SEHRG-DLD technique. The performance analysis of the SEHRG-DLD technique is examined by employing the ChatGPT-generated dataset. The experimental results clearly stated the supremacy of the SEHRG-DLD technique over other recent methods for different measures.
Early detection of chronic kidney disease using eurygasters optimization algorithm with ensemble deep learning approach Sulima M. Awad Yousif, Hanan T. Halawani, Ghada Amoudi, Fathea M. Osman Birkea, Arwa M.R. Almunajam, Azhari A. Elhag Alexandria Engineering Journal, 2024 Chronic kidney disease (CKD) emerges as a global health problem with high morbidity and mortality rates, and it induces other diseases. Patients often fail to notice these diseases because there are no obvious symptoms during the earlier stage of CKD. Earlier diagnosis of CKD allows them to receive early medical intervention to improve the disease progression and outcome. Earlier stratification of CKD can affect the health care provided to patients through various options, such as kidney transportation, hemodialysis, or pharmacological care in milder cases. Recently, deep learning (DL) and machine learning (ML) approaches have gained significance in the domain of medical diagnoses due to their high prediction accuracy. DL and ML approaches can successfully aid physicians in achieving these goals due to their accurate and fast detection performance. The performance of the proposed technique mainly depends on selecting the suitable algorithms and appropriate features. Thus, the study introduces an Eurygasters Optimization Algorithm with an Ensemble DL CKD detection (EOAEDL-CKDD) method. The EOAEDL-CKDD method aims to detect and classify the presence of CKD using feature selection and hyperparameter tuning strategies. The EOAEDL-CKDD method applies a min-max scalar to convert the data into a uniform format to achieve this. The EOAEDL-CKDD technique exploits EOA to select features. This is followed by an ensemble of long short-term memory (LSTM), bidirectional gated recurrent unit (BiGRU), and bidirectional LSTM (BiLSTM) models that are used for the CKD detection process. Finally, a shuffled frog leap algorithm (SFLA) based hyperparameter selection process is carried out to choose the ensemble models' hyperparameter values optimally. The empirical assessment of the EOAEDL-CKDD method is tested on the benchmark CKD dataset. The experimental values highlighted that the EOAEDL-CKDD technique gains an optimal detection rate compared to existing models.
Using Machine Learning for Non-Invasive Detection of Kidney Stones Based on Laboratory Test Results: A Case Study from a Saudi Arabian Hospital Hanan Alghamdi, Ghada Amoudi Diagnostics, 2024 Kidney stone disease is a widespread urological disorder affecting millions globally. Timely diagnosis is crucial to avoid severe complications. Traditionally, renal stones are detected using computed tomography (CT), which, despite its effectiveness, is costly, resource-intensive, exposes patients to unnecessary radiation, and often results in delays due to radiology report wait times. This study presents a novel approach leveraging machine learning to detect renal stones early using routine laboratory test results. We utilized an extensive dataset comprising 2156 patient records from a Saudi Arabian hospital, featuring 15 attributes with challenges such as missing data and class imbalance. We evaluated various machine learning algorithms and imputation methods, including single and multiple imputations, as well as oversampling and undersampling techniques. Our results demonstrate that ensemble tree-based classifiers, specifically random forest (RF) and extra tree classifiers (ETree), outperform others with remarkable accuracy rates of 99%, recall rates of 98%, and F1 scores of 99% for RF, and 92% for ETree. This study underscores the potential of non-invasive, cost-effective laboratory tests for renal stone detection, promoting prompt and improved medical support.
Blockchain with optimal deep learning assisted secure data sharing and classification on future healthcare systems Adwan A. Alanazi, Faten Khalid Karim, Sara Abdelwahab Ghorashi, Ghada Amoudi, Saadia Hassan A. Hamza Alexandria Engineering Journal, 2024 In the future of healthcare, Blockchain (BC) technology holds immense potential for improving the security and privacy of data. By allowing the secure and immutable storage of medical files and healthcare-related transactions, BC ensured that sensitive medical data remains tamper-proof and open only to authorized parties. Patients have greater control over their data's development, revoking or granting access as required, but healthcare workers can streamline data sharing and ensure the integrity of important data. The decentralized nature of BC networks decreases the risk of centralized data breaches, eventually fostering trust and transparency in the healthcare ecosystems. Conversely, deep learning (DL) has great to revolutionize healthcare diagnostics in the future, offering quick and extremely accurate estimates of medical conditions. This technology has greatly enhanced patient solutions, decreased medical expenses, and improved the burden on medical staff by providing appreciated insights into an extensive range of conditions, from cancer to neurological disorders. With this stimulus, this study presents a novel BC with optimal DL-based secure data sharing and classification (BCODL-SDSC) technique in the future healthcare system. The goal of the BCODL-SDSC technique is to secure and thoroughly examine healthcare data using BC and DL techniques. Primarily, the BCODL-SDSC technique enables BC technology to store and maintain the patient’s data from the procedure of several transactions and enable access control to the various stakeholders. For the security of the medical images, the BCODL-SDSC technique applies the Fractional Order Lorenz system (FOLS) based encryption technique with tuna swarm optimization (TSO) algorithm based optimal key generation process. Finally, a multi-stage process performs the classification of the medical images: MobileNetv1 feature extractor, artificial rabbit’s optimization (ARO) based hyperparameter tuning, and stacked recurrent neural network (SRNN) based classification. The experimental outcome of the BCODL-SDSC technique was examined on a benchmark medical image database. An extensive comparative study reported that the BCODL-SDSC technique reaches an effective performance with other models with a maximum accuracy of 99.11%.
Advancements in Sign Language Recognition: A Comprehensive Review and Future Prospects Bashaer A. Al Abdullah, Ghada A. Amoudi, Hanan S. Alghamdi IEEE Access, 2024 Sign language (SL) is a vital mode of communication, bridging the gap between the hearing impaired and hearing communities. However, SL, despite its paramount importance, has received relatively limited attention from researchers. Its unique structural characteristics, distinct from those of natural languages, present novel challenges that require innovative solutions. Remarkable technological advancements, notably in Artificial Intelligence (AI) and machine learning, offer promising avenues for automated Sign Language translation Systems (SLTS). This review study addresses the crucial need for a comprehensive synthesis of existing research by systematically examining and evaluating the progress made in SLTS. By analyzing 58 research papers, with a particular emphasis on the most frequently cited papers from each year up to 2023, we shed light on the field’s current state, identifying key advancements and challenges. This review followed a systematic approach based on clear guidelines. The methodology involved defining research questions, formulating queries, selecting studies based on clear criteria, and extracting pertinent information to address the research objectives. This review found that deep learning techniques, such as convolutional and recurrent neural networks, have shown high accuracy in sign language recognition, and their performance in recognizing the variety of signs has steadily improved over time. Additionally, integrating non-manual features has proven pivotal in enhancing recognition accuracy. Future research should refine advanced deep learning models and integrate non-manual features to improve system accuracy and applicability. These ongoing advancements hold the potential to revolutionize communication and break down barriers for individuals who rely on sign language as their primary mode of communication.
