Explore the vast range of titles available.

Background Drug response prediction is critical in precision medicine to determine the most effective and safe treatments for individual patients. Traditional prediction methods relying on demographic and genetic data often fall short in accuracy and robustness. Recent graph-based models, while promising, frequently neglect the critical role of atomic interactions and fail to integrate drug fingerprints with SMILES for comprehensive molecular graph construction. Results We introduce multimodal multi-channel graph attention network with adaptive fusion (MGATAF), a framework designed to enhance drug response predictions by capturing both local and global interactions among graph nodes. MGATAF improves drug representation by integrating SMILES and fingerprints, resulting in more precise predictions of drug effects. The methodology involves constructing multimodal molecular graphs, employing multi-channel graph attention networks to capture diverse interactions, and using adaptive fusion to integrate these interactions at multiple abstraction levels. Empirical results demonstrate MGATAF’s superior performance compared to traditional and other graph-based techniques. For example, on the GDSC dataset, MGATAF achieved a 5.12% improvement in the Pearson correlation coefficient (PCC), reaching 0.9312 with an RMSE of 0.0225. Similarly, in new cell-line tests, MGATAF outperformed baselines with a PCC of 0.8536 and an RMSE of 0.0321 on the GDSC dataset, and a PCC of 0.7364 with an RMSE of 0.0531 on the CCLE dataset. Conclusions MGATAF significantly advances drug response prediction by effectively integrating multiple molecular data types and capturing complex interactions. This framework enhances prediction accuracy and offers a robust tool for personalized medicine, potentially leading to more effective and safer treatments for patients. Future research can expand on this work by exploring additional data modalities and refining the adaptive fusion mechanisms.

Renal cell carcinomas (RCCs) are the seventh most widespread histological cancer. Around 40% of patients die in RCC due to the disease development. Thus, this tumour is the most lethal malignant urological tumour. The histopathologic classification of RCC is vital for the prognosis, diagnosis, and patient management. Classification and detection of intricate RCC histologic patterns on surgical and biopsy surgery slides under a microscope endures a comprehensively specified, time-consuming task and error-prone for pathologists. A wholly automatic and accurate technique of grading kidney tumours from histopathology images (HIs) is in great demand for recognizing harmful cancers. The correct classification of RCC stage and grade is vital for managing medical management, prognosis, and molecular-based treatments. Many preceding works concentrate on machine learning (ML) and deep learning (DL) methods for the RCC classification. The application of DL to study the histopathological images of kidneys, breasts, etc., and other organs contains several tasks like classification of cancer subtypes and grading. This study presents a Computer Vision Assisted Deep Transfer Learning Model for the Accurate Grading of the RCC (CVDTLM-AGRCC) technique. The CVDTLM-AGRCC technique enables the detection and classification of RCC from kidney histopathology images. Initially, the CVDTLM-AGRCC technique applies the image pre-processing stage using a Gaussian filter (GF) to prevent and eliminate the noise. Furthermore, the fusion of ShuffeNetV2-1.0-SE and CapsNet models is employed for the feature extraction. Moreover, the CVDTLM-AGRCC method uses a hybrid of convolutional neural networks and bidirectional long short-term memory (CNN-BiLSTM) techniques for the RCC classification. Finally, the crayfish optimization algorithm (COA) is used for the hyperparameter tuning of the CNN-BiLSTM method. The efficiency of the CVDTLM-AGRCC approach is examined under the KMC dataset. The comparison study of the CVDTLM-AGRCC approach portrayed a superior accuracy value of 93.89% over existing techniques.

Malicious domains are one of the main resources mandatory for adversaries to run attacks over the Internet. Owing to the significant part of the domain name system (DNS), detailed research has been performed to detect malicious fields according to their unique behaviour, which is considered in dissimilar stages of the DNS life cycle queries and explanations. The DNS has played a crucial role in the evolution of the Internet. Its primary objective is to simplify user experience by converting a website’s Internet Protocol (IP) address into a recognizable domain name and vice versa. Identifying these adverse fields is meaningful in contesting increased network attacks. Artificial intelligence (AI) is applied to develop the areas of malicious domain recognition and hindrance by the probability to improve robust, efficient, and scalable malware detection units. AI methods have expressed significant results in malicious domain detection. This manuscript presents an Enhance Malicious Domain Detection Using an Attention-Based Deep Learning Model with Optimization Algorithms (EMDD-ADLMOA) technique. The proposed EMDD-ADLMOA technique relies on improving malicious domain detection in cybersecurity. Initially, the min–max scaling method is utilized in the pre-processing phase to convert input data into an appropriate design. For feature selection (FS), the proposed EMDD-ADLMOA technique utilizes the quantum-inspired firefly algorithm (QIFA) model. Furthermore, the hybrid model of a temporal convolutional network and bi-directional long short-term memory with squeeze-and-excitation Attention (TCN-BiLSTM-SEA) model is employed for the classification process. Finally, the parrot optimization (PO) model optimally fine-tunes the hyperparameter values of the TCN-BiLSTM-SEA model. The performance results of the EMDD-ADLMOA approach are verified under a malicious dataset. The experimental validation of the EMDD-ADLMOA approach portrayed a superior accuracy value of 98.52% over existing techniques.

Bone marrow (BM) plays a crucial role in the hematopoietic process, producing all of the body’s blood cells and maintaining the overall immune and health system. Red and yellow BM are the two various kinds of BM. A comprehensive identification of these cells assists in the primary and precise recognition of these disorders. The recognition and identification of BM cells are crucial bases for haematology diagnostics. Physical study of BM detection and classification presently performed in medical laboratories can be primarily insufficient owing to various factors, such as prolonged and challenging. Recently, with the fast growth of deep learning (DL) and machine learning (ML) methods, object detection methods have been progressively used for cell detection. DL is a secondary domain of artificial intelligence (AI) methods able to spontaneously assess delicate graphical features to create exact predictions that have been newly popularized in various imaging-related tasks. This study proposes a Multimodal Transfer Learning with Snake Optimization on Bone Marrow Cell Classification (MTLSO-BMCC) technique using biomedical histopathological images. The main intention of the MTLSO-BMCC technique is to identify and classify BM cells utilizing HI. To achieve this, the presented MTLSO-BMCC method initially performs image preprocessing using a median filter (MF) for noise removal. Besides, the multimodal feature extraction process is accomplished in InceptionV3, Deep SqueezeNet, and SE-DenseNet models. The presented MTLSO-BMCC technique employs the hybrid kernel extreme learning machine (HKELM) method for the BM classification method. Finally, the snake optimization algorithm (SOA) is implemented to tune the parameter of the HKELM model. A widespread MTLSO-BMCC methodology simulation is accomplished under the BM Cell Classification dataset. The experimental validation of the MTLSO-BMCC methodology portrayed a superior accuracy value of 98.60% over existing approaches.

Denial of Wallet (DoW) attacks are a cyber threat designed to utilize and deplete an organization’s financial resources by generating excessive prices or charges in their cloud computing (CC) and serverless computing platforms. These threats are primarily appropriate in serverless manners because of features such as auto-scaling, pay-as-you-go, restricted control, and cost growth. Serverless computing, frequently recognized as Function-as-a-Service (FaaS), is a CC method that permits designers to construct and run uses without the requirement to accomplish typical server structure. Detecting DoW threats involves monitoring and analyzing the system-level resource consumption of specific bare-metal mechanisms. Efficient and precise detection of internal DoW threats remains a crucial challenge. Timely recognition is significant in preventing potential damage, as DoW attacks exploit the financial model of serverless environments, impacting the cost structure and operational integrity of services. In this study, a Multi-Head Attention-based Recurrent Neural Network for Denial of Wallet Attacks Detection (MHARNN-DoWAD) technique is developed. The MHARNN-DoWAD method enables the detection of DoW attacks on serverless computing environments. At first, the presented MHARNN-DoWAD model performs data preprocessing by using min-max normalization to convert input data into constant format. Next, the wolf pack predation (WPP) method is employed for feature selection. The detection and classification of DoW attacks, the multi-head attention-based bi-directional gated recurrent unit (MHA-BiGRU) model is utilized. Eventually, the improved secretary bird optimizer algorithm (ISBOA)-based hyperparameter choice process is accomplished to optimize the detection results of the MHA-BiGRU model. A comprehensive set of simulations was conducted to demonstrate the promising results of the MHARNN-DoWAD method. The experimental validation of the MHARNN-DoWAD technique portrayed a superior accuracy value of 98.30% over existing models.

This paper investigates an intelligent reflecting surface (IRS)-assisted hybrid high-altitude platform (HAP) and unmanned aerial vehicle (UAV) uplink communication network employing non-orthogonal multiple access (NOMA). A unified uplink system model is developed that jointly captures UAV three-dimensional deployment, IRS-assisted composite channels, SIC-based uplink NOMA reception, and shared-spectrum cross-tier interference between HAP and UAV layers. To efficiently support dense uplink connectivity, a joint uplink sum-rate maximization problem is formulated under practical mobility, power, quality-of-service, and interference constraints, resulting in a highly non-convex mixed-integer optimization problem. A low-complexity block coordinate descent (BCD)-based framework is proposed to iteratively optimize UAV deployment, user association, uplink power allocation, and IRS phase shifts. The proposed framework is particularly well-suited for high-capacity data offloading and IoT-based safety monitoring in remote mining environments, supporting the digital transformation of the mining sector in alignment with Saudi Vision 2030. Extensive simulations demonstrate that the proposed framework significantly outperforms conventional aerial and terrestrial benchmarks. Specifically, compared to hybrid HAP–UAV uplink NOMA without IRS, the proposed design improves the uplink sum rate by up to 23.03% and by an average of 15.39% over K=10–100 simultaneously scheduled users. Relative to UAV-only and HAP-only IRS-assisted uplink schemes, gains of up to 43.80% and 59.01% are achieved, respectively, while effective improvement is observed over terrestrial uplink baselines under dense user loading. Moreover, IRS-assisted user association yields additional gains of up to 21.45% in sum rate, while outage probability is substantially reduced across the entire SINR range. These results confirm that IRS-assisted hybrid HAP–UAV uplink NOMA provides a scalable and reliable solution for future 6G-oriented dense uplink communication networks.

Tuberculosis (TB) is a chronic lung disorder caused by bacterial infection and is a major cause of death. Lung cancer also has a significant impact, and existing solutions concentrate on initial screening, which mainly results in better outcomes at a comparatively lower cost. Screening, particularly by chest X-rays (CXR), is globally recognized as an effective method for reducing lung cancer mortality. Therefore, a precise and initial identification of TB is highly crucial, or else, it threatens lives. In the investigation into cases of TB, CXR images are not only the primary method of diagnosis according to medical imaging, but also the radiological diagnosis. The recent developments of computing, deep learning (DL), for image processing, carry a beneficial effect for the automated identification of numerous illnesses from CXRs. Now, the effectiveness of lung segmentation and TB screening methods is established for CXRs analysis by the DL technique to help radiologists recognize suspicious lesions and nodes in lung cancer patients. This paper presents an Efficient Deep Learning-Based Hierarchical Feature Fusion Approach for Lung Segmentation and Tuberculosis Screening (EDLHFFA-LSTS) model. The aim is to develop an automatic DL-based framework for precise lung segmentation and TB screening using CXR images to support early diagnosis and clinical decision-making. Initially, the image pre-processing stage includes resizing, adaptive filtering (AF), and histogram equalization (HE) to enhance the image quality. For the segmentation process, the EDLHFFA-LSTS model implements the Res-UNet method. Furthermore, the fusion of EfficientNetV2, CapsNet, and Convolutional Vision Transformer (CViT) techniques is employed for the feature extraction process. Finally, the stacked autoencoder (SAE) technique is implemented for classification. Extensive simulations were conducted to demonstrate the promising results of the EDLHFFA-LSTS methodology on the CXR Masks and Labels dataset. The comparison study of the EDLHFFA-LSTS methodology illustrated a superior accuracy value of 98.33% over existing models.

This study delves into the effects of crude oil and gas prices on the United States’ (US) conventional, Islamic, and environmental, social, and governance (ESG) stock indices from January 2013 to December 2022. Decomposing original time series data to minimise inherent fluctuations and using the Quantile-on-Quantile (QQ) regression approach presents a nuanced view of how these energy prices impact different stock indices. The findings reveal that crude oil prices have a variable impact on the indices: high prices negatively influence the indices, low prices have a positive effect, and moderate prices yield a moderate positive impact. After data decomposition, this positive influence diminishes in higher quantiles, indicating an emerging neutral effect in stabilised conditions. In contrast, gas prices show a limited impact, with high prices slightly benefiting conventional and ESG indices but less so for the Islamic index. This suggests a more pronounced influence of oil prices on the indices, likely due to the dependence of many listed companies on oil. The study emphasises the importance of considering oil-related risks in investment strategies and highlights the asymmetric impact of crude oil prices on the US stock indices. These findings have significant implications for investors and policymakers. They underscore the need for careful consideration of oil price dynamics in investment decisions and the importance of staying vigilant against shifts in oil prices that could lead to market instability.

Skin diseases, ranging from benign conditions to malignant tumors such as melanoma, present substantial diagnostic challenges due to their visual complexity and the inherent subjectivity in manual examination. This paper introduces a hybrid deep learning framework specifically designed for skin lesion segmentation and multi-class classification using dermoscopic images. The proposed model integrates a dual-task architecture, which combines a U-Net-based segmentation network with a classification module based on the EfficientNet-B0 backbone. To improve model interpretability and foster clinical trust, Grad-CAM is incorporated, allowing clinicians to visualize heatmaps that highlight the regions influencing the model's decisions. The model was trained and evaluated on the HAM10000 dataset, demonstrating robust performance, with a Dice coefficient surpassing 0.85 for segmentation and classification accuracy nearing 85%. Despite challenges such as class imbalance and the variety of lesion types, the model provides reliable results across different skin conditions. The use of explainable AI (XAI) enhances transparency, a crucial factor in the clinical acceptance of AI-based diagnostic tools. This approach shows promise in improving diagnostic accuracy and supporting dermatologists, especially in resource-constrained settings, by providing both accurate lesion delineation and reliable class predictions. Future research will focus on improving the model's generalizability, addressing underrepresented classes, and validating its effectiveness in real-world scenarios.

Cyber-physical systems (CPSs) in autonomous vehicles must handle highly dynamic and uncertain settings, where unanticipated impediments, shifting traffic conditions, and environmental changes all provide substantial decision-making issues. Deep reinforcement learning (DRL) has emerged as a strong tool for dealing with such uncertainty, yet current DRL models struggle to ensure safety and optimal behaviour in indeterminate settings due to the difficulties of understanding dynamic reward systems. To address these constraints, this study incorporates double deep Q networks (DDQN) to improve the agent’s adaptability under uncertain driving conditions. A structured reward system is established to accommodate real-time fluctuations, resulting in safer and more efficient decision-making. The study acknowledges the technological limitations of automobile CPSs and investigates hardware acceleration as a potential remedy in addition to algorithmic enhancements. Because of their post-manufacturing adaptability, parallel processing capabilities, and reconfigurability, field programmable gate arrays (FPGAs) are used to execute reinforcement learning in real-time. Using essential parameters, including collision rate, behaviour similarity, travel distance, speed control, total rewards, and timesteps, the suggested method is thoroughly tested in the TORCS Racing Simulator. The findings show that combining FPGA-based hardware acceleration with DDQN successfully improves computational efficiency and decision-making reliability, tackling significant issues brought on by uncertainty in autonomous driving CPSs. In addition to advancing reinforcement learning applications in CPSs, this work opens up possibilities for future investigations into real-world generalisation, adaptive reward mechanisms, and scalable hardware implementations to further reduce uncertainty in autonomous systems.