Ahmed Ibrahim Taloba

Neonatal mortality is still a big health problem worldwide, and in intensive care units, prompt identification of high-risk infants is important to enhance clinical outcomes. Most of the prediction methods used today may fail to accurately capture physiological data that is both heterogeneous and measured at irregular intervals over time and/or varies between medical and healthcare institutions, potentially compromising their reliability and scalability. To mitigate these issues, the study suggests a novel framework for neonatal risk prediction using explainable AI in a cloud-based environment that integrates cutting-edge temporal learning with explainable analytics in an intelligent healthcare setting. A combination of a Temporal Fusion Transformer (TFT) and a Neural Controlled Differential Equation (Neural CDE) network is used to model the long-range temporal dependence and continuous-time physiological dynamics available in neonatal monitoring data. The static clinical features and time-varying physiological signals are encoded temporally in two parallel processors, which are fused together by a temporal invariant fusion mechanism, to obtain detailed representations for patients. A feature-attribution module, which is a part of the Explainable Artificial Intelligence (XAI) module, further increases transparency and aids clinician trust in model predictions. The framework is cloud-enabled, which allows for scalable processing, real-time monitoring, and intelligent decision support in healthcare institutions. The very high AUC-ROC of prediction for mortality (98.63%) and for respiratory failure (98.74%) in an experimental evaluation of a multi-center neonatal intensive care dataset of patients showed good predictive performance, and length of stay was correlated with the model's predictions. The results show the effectiveness, interpretability, and generalizability of the framework to assist in proactive health care management of neonates.

Worldwide stroke is still among the top ten causes of death and permanent disability, thus the importance of early and reliable risk prediction systems. Current machine learning (ML) models in healthcare, however, have been plagued with issues related to data privacy leakage, algorithmic bias, and lack of interpretability, and have failed to reach clinical use. To predict stroke risk securely in privacy-sensitive clinical settings, this study proposes a Fairness-Aware and Trustworthy Machine Learning Framework (FAT-MLF). This framework combines privacy-preserving data management with differential privacy, fairness-oriented optimization and fairness constraints, such as demographic parity, and explainable AI with interpretability, such as SHAP. To capture the nonlinear interactions between clinical features, a hybrid predictive model using the Gradient Boosting Machine (GBM) and deep neural networks is used. The experimental results on the stroke benchmark dataset with 5000 patient data records show the proposed model with an accuracy of 96.8%, a precision of 95.4%, a recall of 94.9%, and an F1-score of 95.1%, which reduces the demographic bias by 32% compared to baseline ML models. Under ε-differential privacy constraints, privacy analysis verifies that the data is resistant to a data reconstruction attack. The findings suggest that embedding fairness, privacy, and trustworthiness substantially enhances predictive accuracy and ethical adherence of clinical AI systems. The framework, as proposed, is scalable and can be implemented in real-world healthcare systems to safely and fairly predict stroke risk.

The fast computerization of health care has created the issue of data leakage, unauthorized access and patient loss of trust. As Electronic Health Records (EHRs) become the focus of clinical processes, data protection, security, and auditing have become a matter of concern. Conventional centralized medical records are still vulnerable to single points of vulnerability, untraceability, and misuse of privileges, resulting in data breach and lack of trust of security in health care data management. To mitigate these problems, this study will propose a privacy-sensitive medical data management system based on blockchain technology to offer secure, transparent, and role-based access to EHRs. The framework combines multi-layer cryptographic security, decentralized authorization through smart contracts, and encrypted storage with the InterPlanetary File System (IPFS) to secure confidentiality and scalability. Access transactions are stored verifiably and tamper-resistant on the blockchain, which does not affect efficiency. Based on the UCI Heart Disease data, the framework was able to attain 98.5% access success, 220 access requests/sec, and 96% privacy retention rates, which is 12–15 times higher than traditional centralized EHR systems. These findings illustrate a robust, scalable and privacy ensuring alternative that guarantees integrity, accountability and efficiency of operations in medical data sharing.

The integration of gamification in education has emerged as a transformative approach to enhance student engagement and academic achievement. Traditional language learning methods often lack personalization, real-time feedback, and adaptive assessment, leading to disengagement and ineffective learning. Existing AI-driven language learning approaches, while promising, primarily focus on rule-based assessments or simple feedback mechanisms, failing to provide context-aware, interactive, and adaptive learning experiences. To address these limitations, this study proposes a gamification-enhanced, Deep learning-driven Transformer framework that offers intelligent feedback and adaptive assessment to enhance language learning outcomes. The gamification elements such as leaderboards, achievement badges, adaptive difficulty levels, and interactive challenges enhance engagement and motivation. The adaptive assessment mechanism dynamically adjusts task complexity based on the learner’s progress, ensuring a customized learning experience. The proposed system integrates pre-trained Transformer models to analyze learners' responses in real-time, providing personalized feedback on fluency, grammar, vocabulary, and pronunciation. Cronbach's Alpha was employed to ensure the reliability of the developed questionnaires and achievement tests, confirming their internal consistency. The proposed system achieves 99.9% adaptive feedback accuracy, ensuring precise, real-time assessments that enhance language learning effectiveness. The findings indicate that gamification and advanced Transformer based feedback mechanisms significantly enhance students' comprehension and achievement in education, providing educators with innovative strategies to improve instructional methods and foster a more engaging learning experience. This research highlights the potential of integrating gamification and intelligent assessment technologies to transform education, promoting effective learning outcomes and student success.

Intelligent Speech Technology (IST) is revolutionizing healthcare by enhancing transcription accuracy, disease diagnosis, and medical equipment control in smart hospital environments. This study introduces an innovative approach employing federated learning with Multi-Layer Perceptron (MLP) and Gated Recurrent Unit (GRU) neural networks to improve IST performance. Leveraging the “Medical Speech, Transcription, and Intent” dataset from Kaggle, comprising a variety of speech recordings and corresponding medical symptom labels, noise reduction was applied using a Wiener filter to improve audio quality. Feature extraction through MLP and sequence classification with GRU highlighted the model’s robustness and capacity for detailed medical understanding. The federated learning framework enabled collaborative model training across multiple hospital sites, preserving patient privacy by avoiding raw data exchange. This distributed approach allowed the model to learn from diverse, real-world data while ensuring compliance with strict data protection standards. Through rigorous five-fold cross-validation, the proposed Fed MLP-GRU model demonstrated an accuracy of 98.6%, with consistently high sensitivity and specificity, highlighting its reliable generalization across multiple test conditions. In real-time applications, the model effectively performed medical transcription, provided symptom-based diagnostic insights, and facilitated hands-free control of healthcare equipment, reducing contamination risks and enhancing workflow efficiency. These findings indicate that IST, powered by federated neural networks, can significantly improve healthcare delivery, accuracy in patient diagnosis, and operational efficiency in clinical settings. This research underscores the transformative potential of federated learning and advanced neural networks for addressing pressing challenges in modern healthcare and setting the stage for future innovations in intelligent medical technology.

American scientific publishing group

Driver drowsiness has been identified as a major cause of roadway accidents globally. Efficiently determining the extent of drowsiness can greatly enhance preventive measures. This study proposes a novel approach, combining convolutional neural networks (CNN) and Gated Recurrent Units (GRU) to dynamically evaluate both the presence of drowsiness and its severity based on the Eye Aspect Ratio (EAR). By bridging spatial features extracted by CNNs with temporal sequences through GRU, our model offers a robust and real-time assessment of drowsiness levels, paving the way for enhanced safety measures in vehicular systems. Incorporating Neutrosophic Logic enables a more robust representation of uncertainty and ambiguity in the data and enhances the accuracy of driver drowsiness level detection within the Hybrid CNN-GRU framework. The model’s hybrid CNN-GRU structure combines CNN layers to extract spatial information from Human eye Images and GRU units to represent temporal correlations between frames. In-car cameras and sensors must be integrated to implement the suggested system in real-time and enable continuous driver behavior monitoring. The system alerts early warnings and takes action when drowsiness is detected, lowering the likelihood of accidents caused by weary drivers. The CNN-GRU hybrid architecture accurately detects fatigue during real-time driving. Performance metrics, including accuracy, recall, and F1-score, are provided for comparative research utilizing baseline models. Model behavior may be understood by visualizing tiredness detection and carefully examining false positives and negatives. The proposed CNN-GRU framework outperforms traditional methods such as SVM, KNN, and BPNN by achieving a significantly higher accuracy of 99.5%. It increases the recognition of driver tiredness by proposing a trustworthy and adaptable hybrid CNN-GRU deep learning system. This project is implemented in Python; it offers a practical and versatile solution for real-time driver drowsiness level detection. The proposed technology has the potential to dramatically increase traffic safety by sending out early warnings and taking steps to lessen the risks related to driver fatigue.

American scientific publishing group

American scientific publishing group

The growing dependence on eTextbooks and Massive Open Online Courses (MOOCs) has led to an increase in the amount of students' learning data. By carefully analyzing this data, educators can identify difficult exercises, and evaluate the quality of the exercises when teaching a particular topic. In this study, an analysis of log data from the semester usage of the OpenDSA eTextbook was offered to identify the most difficult data structure course exercises and to evaluate the quality of the course exercises. Our study is based on analyzing students' responses to the course exercises. We applied item response theory (IRT) analysis and a latent trait mode (LTM) to identify the most difficult exercises .To evaluate the quality of the course exercises we applied IRT theory. Our findings showed that the exercises that related to algorithm analysis topics represented the most difficult exercises, and there existing six exercises were classified as poor exercises which could be improved or need some attention.