ABSTRACT This study introduces an intelligent system aimed at improving the safety of school transportation and simplifying attendance tracking through the use of QR codes and Machine Learning techniques. Each student is provided with a unique QR code printed on their identification card, which is scanned when entering and leaving the school bus. This process enables automatic recording of attendance without manual intervention. The system immediately informs parents about their child’s boarding notifications. It also prepares periodic attendance summaries that help school administrators analyze student travel patterns. In addition, GPS-based tracking allows real-time monitoring of bus location, while a Machine Learning model is used to estimate arrival times more accurately. Key Words: QRcode, Attendance, Monitoring, Notification, Tracking, Automation,Bus arrival time prediction.
Melanoma remains one of the most aggressive forms of skin cancer, and early diagnosis is critical to improving patient survival. This study presents an Adaptive Hybrid AI Framework (AHA-Net) designed for accurate and interpretable skin lesion segmentation and melanoma classification. The proposed architecture enhances a modified UNet + + backbone with an Adaptive Scaled Dot Attention Mechanism (A-SDAM) that dynamically regulates attention sharpness across multiple lesion scales. A Residual Cross-Attention Bridge (RCAB) enables effective contextual fusion between segmentation and classification pathways, while Vision Transformer (ViT)–based multi-scale attention layers capture both local and global dependencies. A hybrid CNN–ViT classifier, refined with dynamically weighted SVM post-classification, improves decision boundary precision under class imbalance. Furthermore, self-distillation between ViT layers enhances feature coherence and cross-dataset generalization. The framework was rigorously evaluated across four benchmark datasets i.e., HAM10000, ISIC 2019; ISIC 2020, and PH2, representing diverse lesion types and imaging conditions. Experimental results demonstrate that Proposed AHA-Net consistently outperforms existing state-of-the-art architectures, including U-Net, ResNet, UNet++, and TransUNet, in both segmentation and classification tasks. Statistical analysis confirms the significance and reproducibility of performance gains (p < 0.05). Quantitative explainability assessment using Grad-CAM shows that the model’s attention maps align closely with clinically relevant melanoma features such as irregular borders, color heterogeneity, and asymmetric structures. These results establish Proposed AHA-Net as a robust, generalizable, and explainable AI framework with strong potential for integration into real-world dermatological diagnostic workflows.
In Dentistry, the most common problem that is seen in most of the age groups particularly in children and teenagers are the Occlusal Caries. Accurate identification of occlusal caries is often hindered by the anatomical complexity of posterior teeth and the limited sensitivity of conventional diagnostic approaches. Standard techniques, including visual inspection and radiographic evaluation, may fail to consistently detect incipient lesions. This study proposes a Generative Artificial Intelligencedriven framework for occlusal caries detection based on multimodal odontological imaging. The proposed model combines convolutional neural networks, deep feature extraction mechanisms, and generative learning strategies to fuse heterogeneous imaging inputs for enhanced diagnostic interpretation. The architecture supports real-time analysis, improves precision, and reduces operator-dependent variability. Findings reported in prior research indicate that combines diverse imaging modalities with AI-based analytical models substantially improves detection robustness and reliability. The presented system demonstrates the potential of intelligent multimodal data fusion to advance early diagnosis, optimize clinical decision-making, and contribute to more efficient and technology-based dental care.
Cognitive Radio Internet of Things (CRIoT) networks are vulnerable to jamming attacks because of their vast spectrum usage and limited capabilities of IoT devices. In such attacks, an adversary deliberately transmits interfering signals to disrupt normal communication that causes data loss and potential security risks. Detecting and preventing jamming in CRIoT environments is particularly difficult due to the constantly changing spectrum conditions. In this work, Exponential Cape Lynx Optimizer enabled Dual Horizontal Squash Capsule Network (ECLO_DHS-CapsNet) is devised to detect and mitigate jamming instance in CRIoT. Initially, CRIoT system model is simulated. Here, different types of IoT devices use wireless spectrum to communicate, while a centralized or distributed receiver continuously monitors the incoming signals for any irregularities or unusual activity. When abnormal patterns indicative of interference appears, jamming instance detection is performed using Dual Horizontal Squash Capsule Network (DHS-CapsNet), which is trained by Exponential Cape Lynx Optimizer (ECLO). The ECLO incorporates Exponentially Weighted Moving Average (EWMA) and Cape Lynx Optimizer (CLO). Once it detected the jamming instance, the system performs the mitigation process by blocklisting the malicious users. The ECLO_DHS-CapsNet is evaluated using metrics, such as accuracy, True Positive Rate (TPR), True Negative Rate (TNR), and F1-score, achieving peak values of 96.47%, 96.17%, 95.22%, and 95.21%, respectively.
A numerical research work is carried out to examine fluid-structure interaction in flow over six inline square cylinders undergoing transverse oscillations via the lattice Boltzmann method. The impact of the oscillation frequency ratio ([Formula: see text]), where imposed oscillation frequency is represented as [Formula: see text] and [Formula: see text]is the natural vortex shedding frequency of a stationary cylinder, on wake dynamics and flow patterns is examined. Simulations are performed at Reynolds number of Re = 80, with a non-dimensional gap spacing (s/d = 1.0) and an oscillation amplitude ratio of A/d = 0.2. The frequency ratio is varied in the range 0.1 < [Formula: see text]≤ 2.0. Three distinct flow patterns are identified: synchronous lock-on (0.8 ≤ [Formula: see text]≤ 1.4), quasi-periodic lock-on-I (1.6 ≤ [Formula: see text] ≤ 2.0) and quasi-periodic non-lock-on-I (0.1 ≤ [Formula: see text]≤ 0.6). The synchronization range is wider than that reported for single oscillating cylinder. A single coherent wake envelops the entire cylinder array, with wake recovery occurring at higher oscillation frequencies due to merging of vortex and formation of multi-polar vortices downstream of final cylinder. The first cylinder experiences highest mean drag, followed by a reduction and gradual increase along the downstream cylinders.
