To constructively ameliorate and enhance traffic safety measures in Saudi Arabia, a prolific number of AI (Artificial Intelligence) traffic surveillance technologies have emerged, including Saher, throughout the past years. However, rapidly detecting a vehicle incident can play a cardinal role in ameliorating the response speed of incident management, which in turn minimizes road injuries that have been induced by the accident’s occurrence. To attain a permeating effect in increasing the entailed demand for road traffic security and safety, this paper presents a real-time traffic incident detection and alert system that is based on a computer vision approach. The proposed framework consists of three models, each of which is integrated within a prototype interface to fully visualize the system’s overall architecture. To begin, the vehicle detection and tracking model utilized the YOLOv5 object detector with the DeepSORT tracker to detect and track the vehicles’ movements by allocating a unique identification number (ID) to each vehicle. This model attained a mean average precision (mAP) of 99.2%. Second, a traffic accident and severity classification model attained a mAP of 83.3% while utilizing the YOLOv5 algorithm to accurately detect and classify an accident’s severity level, sending an immediate alert message to the nearest hospital if a severe accident has taken place. Finally, the ResNet152 algorithm was utilized to detect the ignition of a fire following the accident’s occurrence; this model achieved an accuracy rate of 98.9%, with an automated alert being sent to the fire station if this perilous event occurred. This study employed an innovative parallel computing technique for reducing the overall complexity and inference time of the AI-based system to run the proposed system in a concurrent and parallel manner.

Pneumonia, COVID-19, and tuberculosis are some of the most fatal and common lung diseases in the current era. Several approaches have been proposed in the literature for the diagnosis of individual diseases, since each requires a different feature set altogether, but few studies have been proposed for a joint diagnosis. A patient being diagnosed with one disease as negative may be suffering from the other disease, and vice versa. However, since said diseases are related to the lungs, there might be a likelihood of more than one disease being present in the same patient. In this study, a deep learning model that is able to detect the mentioned diseases from the chest X-ray images of patients is proposed. To evaluate the performance of the proposed model, multiple public datasets have been obtained from Kaggle. Consequently, the proposed model achieved 98.72% accuracy for all classes in general and obtained a recall score of 99.66% for Pneumonia, 99.35% for No-findings, 98.10% for Tuberculosis, and 96.27% for COVID-19, respectively. Furthermore, the model was tested using unseen data from the same augmented dataset and was proven to be better than state-of-the-art studies in the literature in terms of accuracy and other metrics.

Over the decades, a tremendous increase has been witnessed in the production of documents available in digital form. The increased production of documents has gained so much momentum that their rate of production jumps two-fold every five years. These articles are searched over the internet via search engines, digital libraries, and citation indexes. However, the retrieval of relevant research papers for user queries is still a pipedream. This is because scientific documents are not indexed based on some subject classification hierarchies. Hence, the classification of these documents becomes a challenging task for the researchers. Classification of the documents can be two-fold: one way is to assign a single label to each document and the other is to assign multi-labels to each document based on its belonging domains. Classification of the documents can be performed by using either the available metadata or the whole content of the documents. While performing classification, there are many challenges which may belong to the dataset, feature selection technique, preprocessing methodology, and which classification model is suitable for the classification of the documents. This paper highlights the issues for single-label and multi-label classification by using either metadata or content of the documents and why metadata-based approaches are better than content-based approaches in terms of feasibility.

Multiple Sclerosis (MS) is characterized by chronic deterioration of the nervous system, mainly the brain and the spinal cord. An individual with MS develops the condition when the immune system begins attacking nerve fibers and the myelin sheathing that covers them, affecting the communication between the brain and the rest of the body and eventually causing permanent damage to the nerve. Patients with MS (pwMS) might experience different symptoms depending on which nerve was damaged and how much damage it has sustained. Currently, there is no cure for MS; however, there are clinical guidelines that help control the disease and its accompanying symptoms. Additionally, no specific laboratory biomarker can precisely identify the presence of MS, leaving specialists with a differential diagnosis that relies on ruling out other possible diseases with similar symptoms. Since the emergence of Machine Learning (ML) in the healthcare industry, it has become an effective tool for uncovering hidden patterns that aid in diagnosing several ailments. Several studies have been conducted to diagnose MS using ML and Deep Learning (DL) models trained using MRI images, achieving promising results. However, complex and expensive diagnostic tools are needed to collect and examine imaging data. Thus, the intention of this study is to implement a cost-effective, clinical data-driven model that is capable of diagnosing pwMS. The dataset was obtained from King Fahad Specialty Hospital (KFSH) in Dammam, Saudi Arabia. Several ML algorithms were compared, namely Support Vector Machine (SVM), Decision Tree (DT), Logistic Regression (LR), Random Forest (RF), Extreme Gradient Boosting (XGBoost), Adaptive Boosting (AdaBoost), and Extra Trees (ET). The results indicated that the ET model outpaced the rest with an accuracy of 94.74%, recall of 97.26%, and precision of 94.67%.

Communication is a crucial aspect of human life and society. It enables us to connect, share ideas, and foster mutual understanding. However, there are individuals who are unable to communicate through spoken words, such as those who are deaf and mute and rely on sign language for communication, especially in education. Unfortunately, sign language is not widely understood, especially in Arabic-speaking countries. Especially, in the emerging cross cultural environment in the Middle eastern countries, this is becoming even challenging for the said individuals. Additionally, poor lighting conditions, the inherent environment noise, and hand obstructions make the detection process even challenging. There are limited studies in literature emphasizing the said issues with cross cultural focus and in a real time situation. To address this challenge, this initiative aims to use computer vision and deep learning to bridge the communication gap and facilitate mutual comprehension with a sustainable solution. The primary goal of this research is to serve Arabic-speaking regions by breaking down communication barriers and enabling seamless communication for the deaf community. After conducting a thorough literature review, the study has shortlisted and investigated CNN and YOLOv8 on the Arabic sign language dataset, as they show promise in solving similar problems. The CNN model achieved an accuracy of 98.88%, whereas the YOLOv8 model achieved precision and recall values of 97%, respectively. These results are quite acceptable and enhanced compared to similar studies in the literature. Moreover, a mobile application has been developed for sign language recognition on the go. Thus, the study is a potential contribution to assist such differently abled people in the Arabic regions.