The article presents a novel method for remote heart rate estimation using a transformer architecture with a multi-skip connection biLSTM decoder. The method leverages skin color variations caused by blood volume changes to estimate heart rate from videos. The proposed framework includes facial region segmentation, feature extraction, and heart rate estimation. Experiments conducted on publicly available datasets and a new in-the-wild dataset demonstrated that the proposed method outperforms existing methods, achieving state-of-the-art results. The study also explores the optimal feature extractor and number of frames for accurate heart rate estimation.

The article presents a computer vision-based approach for the segmentation of chronic subdural hematomas using computed tomography images. The research developed a dataset of CT scans from 21 patients and trained two neural network models based on U-Net architecture. The models were evaluated using a 10-fold cross-validation technique, achieving a Dice metric of 0.77. The solution was integrated into the OsiriX DICOM Viewer, significantly reducing segmentation time and being deemed clinically acceptable by medical experts. The study highlights the potential for AI to enhance medical imaging processes, particularly in neurosurgery.

A novel Maximum Power Point Tracking (MPPT) controller for standalone Wind Energy Conversion Systems (WECS) with Permanent Magnet Synchronous Generators (PMSG) is introduced, featuring an Arbitrary Order Sliding Mode Control (AOSMC) to mitigate measurement noise. The control methodology integrates Feedforward Neural Networks (FFNN) and a high-gain differentiator (HGO) to enhance system performance. Simulation experiments in MATLAB validate the controller's effectiveness in maximizing power extraction. The study emphasizes the importance of renewable energy sources and the advancements in wind energy technology, highlighting the controller's robustness and practical applicability.

The article outlines essential tools for graphic designers, emphasizing the importance of having the right software to enhance workflow, productivity, and work quality. It recommends a variety of tools, including Adobe Creative Cloud Suite, Sketch, Figma, Procreate, Affinity Designer, Canva, Adobe Spark, Pixlr, Adobe XD, and InVision, each praised for their unique features and benefits. The article aims to help designers, both seasoned and new, to navigate the vast array of design tools available and choose the ones that best suit their needs.

The author shares their journey of preparing for and successfully passing the eJPT certification exam on the first attempt. They highlight the importance of practical experience, resourcefulness, and meticulous note-taking. The article details the tools used during the exam and emphasizes the value of persistence and resilience in achieving cybersecurity certifications.

The article introduces a novel three-dimensional conservative chaotic system involving hyperbolic functions, designed to enhance image encryption. The proposed system's dynamical behaviors are analyzed using Lyapunov exponents and bifurcation diagrams. The encryption method employs a modified Diffie–Hellman key exchange protocol to generate a self-invertible key matrix, ensuring high security and resistance to various attacks. The method's effectiveness is demonstrated through simulations, showing robustness against statistical analysis, noise, and other assaults. The study highlights the system's potential for secure multimedia transmission.

Blockchain and AI are transformative technologies that, when combined, offer enhanced security, decentralized infrastructure, and improved data privacy. Practical coding examples using Python, Ethereum, and TensorFlow demonstrate how these technologies can work together to create transparent and privacy-preserving AI systems. The convergence of blockchain and AI is poised to drive innovation and address critical challenges in various domains.

The study investigates the impact of incorporating potassium acetate, calcium acetate, and ethyl acetate as admixtures in concrete to enhance its workability, density, and compressive strength, particularly for structures in Iraqi ports exposed to aggressive marine environments. The research employs a central composite experimental design and response surface methodology (RSM) to optimize and model the properties of concrete mixtures. The findings indicate that calcium acetate is particularly effective, achieving the highest compressive strength and workability improvements. The study concludes that these low-cost acetate admixtures, especially calcium acetate, are promising for fabricating durable and high-strength concrete suitable for marine construction applications.

The study developed machine learning-based predictive models for the tensile and flexural strength of 3D-printed concrete. Data from 25 literature studies were collected, and models such as Gaussian Process Regression, Decision Tree Regression, Support Vector Machine, and XGBoost Regression were used. The Support Vector Machine Regression-based predictive model showed the highest degree of accuracy. The research contributes to the construction industry by enabling efficient selection of optimal ingredients for 3D-printed concrete structures without extensive laboratory trials, potentially improving their structural integrity and safety.

The study utilizes the RUSLE soil loss model combined with remote sensing and GIS technologies to estimate soil erosion and prioritize conservation efforts in seven transboundary sub-watersheds between Iraq and Iran. The findings indicate that soil erosion is more severe in the northern and northeastern parts of the sub-watersheds, primarily in mountainous regions, necessitating high-priority conservation plans. However, the implementation of these plans is challenging due to the terrain. Other sub-watersheds exhibit low soil loss and are given lower priority for conservation. The study emphasizes the importance of slope and rainfall erosivity in influencing soil erosion rates.