Background/Objectives: Alzheimer’s disease is the leading cause of dementia, marked by progressive cognitive decline and a severe socioeconomic burden. Early and accurate diagnosis is crucial to enhancing patient outcomes, yet traditional clinical and imaging assessments are often limited in sensitivity, particularly at early stages. This study presents a dual-modal framework that integrates symptom-based clinical data with magnetic resonance imaging (MRI) using machine learning (ML) and deep learning (DL) models, enhanced by explainable AI (XAI). Methods: Four ML classifiers—K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Decision Tree (DT), and Random Forest (RF)—were trained on demographic and clinical features. For stage-wise classification, five DL models—CNN, EfficientNetB3, DenseNet-121, ResNet-50, and MobileNetV2—were applied to MRI scans. Interpretability was incorporated through SHAP and Grad-CAM visualizations. Results: Random Forest achieves the highest accuracy of 97% on clinical data, while CNN achieves the best overall performance of 94% in MRI-based staging. SHAP and Grad-CAM were used to find clinically relevant characteristics and brain areas, including hippocampal atrophy and ventricular enlargement. Conclusions: Integrating clinical and imaging data and interpretable AI improves the accuracy and reliability of AD staging. The proposed model offers a valid and clear diagnostic route, which can assist clinicians in making timely diagnoses and adjusting individual treatment.

Driver drowsiness and fatigue are leading causes of road accidents, particularly due to drivers having to operate vehicles for long hours under challenging phys Driver drowsiness and fatigue are leading causes of road accidents, particularly due to drivers having to operate vehicles for long hours under challenging phys

Background/Objectives: The human brain regulates physiological processes, cognitive functions, and emotional responses. Brain-related conditions, such as neurological disorders, strokes, and tumors, are complex and pose significant challenges to medical professionals. Among these, brain tumors are particularly critical due to their impact on essential functions and the difficulty in achieving accurate diagnosis and classification. This study aims to explore the application of deep learning models in brain tumor image classification, focusing on improving diagnostic accuracy through model optimization. Methods: This research conducts a comparative analysis of four deep learning architectures: CNN, VGG-19, Inception V3, and ResNet-10. Each model's performance is evaluated on validation accuracy over ten epochs, both with and without hyper parameter tuning. Key hyper parameters, such as learning rate and optimizer selection, are adjusted to enhance model performance. Results: The CNN achieved a baseline validation accuracy of 77.86%, which improved to 89.31% after hyper parameter tuning. VGG-19, with tuning, reached a validation accuracy of 70.23%. ResNet-10 performed the worst, maintaining an accuracy of 51.91%, even with tuning. Inception V3 showed moderate performance, achieving a validation accuracy of 55.73%. The results highlight the significant impact of hyper parameter optimization on model accuracy. Conclusions: Fine-tuning hyper parameters and selecting appropriate models are critical to improving the accuracy of brain tumor classification. These findings provide insights into developing practical and efficient deep learning models, paving the way for advancements in diagnostic imaging, early detection, and clinical neuroscience.

Background and aim: Breast cancer remains the most common malignancy and the leading cause of cancer-related death among women globally, especially in low- and middle-income countries. Early and accurate diagnosis is critical to improving survival outcomes. This study compares the diagnostic accuracy of ultrasonography and mammography in detecting breast carcinoma, using histopathology as the gold standard. Methods: A cross-sectional descriptive study was conducted in a radiology department at a tertiary care hospital, including 240 female patients presenting with clinically palpable breast lesions. Each participant underwent both mammography and ultrasonography prior to biopsy. Data collection included lesion characteristics, imaging findings, and histopathological results. Diagnostic performance was evaluated using sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and overall accuracy. Results: Ultrasound demonstrated a consistent diagnostic accuracy of 86.3% in both breasts, with high NPVs (≥95.7%) and good sensitivity in dense tissue. Mammography showed variability between sides: accuracy was 81.7% on the right and 94.6% on the left, with left-sided sensitivity reaching 100%. Ultrasound was more effective in identifying hyperechoic lesions and smooth margins, while mammography was superior in detecting calcifications and architectural distortions. Statistical testing confirmed significant associations ( P < 0.001) for all modality comparisons with histopathology. Conclusion: This study highlights the complementary diagnostic value of ultrasonography and mammography in the evaluation of breast carcinoma. Ultrasound demonstrated consistent diagnostic accuracy and strong NPV, making it particularly valuable in premenopausal women and those with dense breast tissue. Mammography, on the other hand, showed superiority in detecting microcalcifications and subtle architectural distortions, underscoring its indispensable role in identifying ductal carcinoma in situ and early-stage malignancies.

Background/Objectives: The human brain is responsible for controlling various physiological functions, and hence, the presence of tumors in the brain is a major concern in the medical field. The correct identification and categorization of tumors in the brain using Magnetic Resonance Imaging (MRI) is a major requirement for the diagnosis and treatment of a tumor. The proposed research will focus on designing a CNN model that is optimized for tumor image classification. Methods: This research proposes an optimized CNN model featuring strategically placed dropout layers and hyperparameter optimization. This study uses a dataset of 640 MRI scans (320 tumor and 320 non-tumor) collected from a private hospital in Saudi Arabia. The proposed method utilizes a learning rate of 0.001 in combination with the Adam optimizer to ensure stable and efficient convergence. Its performance was benchmarked against established architectures, including VGG-19, Inception V3, ResNet-10, and ResNet-50, with evaluation based on classification accuracy and computational cost. Results: The experimental results show that the optimized CNN proposed in this work performs much better than the deeper architectures. The network reached a maximum training accuracy of 97.77% and a final test accuracy of 95.35% with a small test loss of 0.2223. The test accuracy of the optimized VGG-19 and Inception V3 networks was much lower, with a training time per epoch that was several orders of magnitude higher. The validation stability of the proposed network was high (92.25% to 95.35%) during the final stages of training. Conclusions: The conclusion drawn from this study is that hyperparameter optimization and strategic regularization are more advantageous for tumor classification using MRI images than the mere depth of the model. The accuracy of 95.35% with low computational complexity makes this lightweight CNN model a feasible solution for real-time applications.