Mesopotamian Journal of Computer Science

An Extensive Examination of the IoT and Blockchain Technologies in Relation to their Applications in the Healthcare Industry

Karthik Kumar Vaigandla, Madhu Kumar Vanteru, Mounika Siluveru — Sun, 14 Jan 2024 00:00:00 +0000

Numerous domains have been transformed by the communication technologies made possible by the Internet of Things (IoT), one of which is health monitoring systems. Patterns associated with diseases and health conditions can be identified through the utilization of machine learning and cutting-edge AI techniques. Currently, scientific endeavours are concentrated on enhancing IoT-enabled applications such as medical report administration, prescription traceability, and infectious disease surveillance through the amalgamation of blockchain technology(BCT) and machine learning(ML) models. Although recent advancements have attempted to increase the adaptability of blockchain(BC) and ML for IoT applications, there are still a number of crucial considerations that must be made for improved outcomes. This report provides a comprehensive examination of emerging technologies in the healthcare sector, encompassing the IoT, and blockchain.

New Recruitment Approach Based on Logistic Regression Model

Ishraq Hatif Abd Almajed, Ghalia Nassreddine, Joumana Younis — Sun, 18 Feb 2024 00:00:00 +0000

Artificial intelligence (AI) is a pivotal technological advancement developed by humans with the aim of enhancing the quality of human existence. It signifies the capacity of a computerized machine resembling a robot to execute tasks typically performed by humans and replicate human behavior. Machine learning (ML), a subfield of AI, involves the construction of systems that acquire the ability to make predictions about new output values by leveraging existing data, without the need for human interaction. Currently, ML has been incorporated into various fields, including but not limited to medical diagnosis, image processing, prediction, classification, learning association, commerce, finance, and natural language processing. This project aims to employ ML techniques within the human resource (HR) department. The implemented model will enable the human resources department to effectively identify the most appropriate candidates for a job opening throughout the recruitment process, utilizing a comprehensive dataset and considering many criteria, all without the need for manual intervention. The construction of the model involves the utilization of an authentic dataset comprising recruitment tasks. Initially, the dataset undergoes a process of selecting the most pertinent elements from both pre-existing and extracted factors. These selected factors include educational level, age, and past experience. Furthermore, taking into consideration these aforementioned factors, a decision system is constructed utilizing the Binary classification approach. The logistic regression classifier is utilized in this investigation. Subsequently, the dataset is partitioned into two distinct subsets, namely the training subset and the testing subset. The effectiveness of the model is demonstrated by the utilization of various evaluation metrics, including the confusion matrix, recall, precision, accuracy, and F-measure values.

Examining Ghana's National Health Insurance Act, 2003 (Act 650) to Improve Accessibility of Artificial Intelligence Therapies and Address Compensation Issues in Cases of Medical Negligence

George Benneh Mensah, Maad M. Mijwil, Mostafa Abotaleb — Sat, 02 Mar 2024 00:00:00 +0000

Objective: Examine Ghana’s National Health Insurance Act (Act 650) to identify coverage gaps limiting artificial intelligence (AI) therapy access and address medical negligence liability issues surrounding automated healthcare systems.

Methods: Legal and regulatory analysis of Act 650 were conducted, review of academic literature on global uptake of AI interventions and medical negligence principles were elucidated, examination of case studies implementing pilot AI therapy programs under insurance schemes were considered.

Results & Conclusions: Act 650 lacks clear provisions for funding innovative AI treatments with proven efficacy and undefined negligence determination guidelines involving AI systems, contributing to accessibility and accountability issues. Proposed amendments to reimburse certain AI therapies through the National Health Insurance Scheme, expand certified provider eligibility, and institute transparent negligence compensation formulas.

Recommendations: Reform Act 650 to support increased appropriate use of AI healthcare services, protect patients undergoing automated diagnosis/treatment, and clarify liability rules for medical negligence incidents relating to AI.

Novelty & Significance: First extensive analysis focused on opportunities for Ghana’s health insurance framework to catalyze equitable diffusion of advanced AI therapeutics and address emerging legal challenges and safety risks as automated medicine advances.

A Comparative study of Chest Radiographs and Detection of The Covid 19 Virus Using Machine Learning Algorithm

Shaimaa Q. Sabri, Jahwar Y. Arif, Ghada A. Taqa, Ahmet Çınar — Mon, 11 Mar 2024 00:00:00 +0000

The severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) outbreak that is causing coronavirus disease 2019 is being deemed a pandemic because of its quick spread around the globe. Because chest X-ray pictures have shown to be beneficial in monitoring a variety of lung disorders, they have recently been utilized to monitor COVID-19 disease. It takes time to manually analyze a lot of chest X-ray pictures. Several previous studies have suggested machine-learning (ML)-based techniques for COVID-19 detection from chest X-ray pictures as a solution to this issue. Though little effort has been made to use traditional machine learning (ML) methods, the majority of these investigations use deep learning (DL) based techniques. Conventional ML-based algorithms will be favored for implementation if they can yield identical outcomes as DL-based methods. In this effort, we constructed four classic ML-based models for COVID-19 identification, driven by the need to close the gap in the literature. The accuracy rates for the various classification models were as follows, according to the results: 93.4% for Support Vector Machine (SVM), 93.3% for Random Forest (RF), 90.5% for K-Nearest Neighbors (KNN), and 87.9% for Decision Tree (DT). The results of the study showed that machine learning-based algorithms can produce great results for COVID-19 identification by being refined and improved using several well-known data preparation approaches.

Advanced Image Processing Techniques for Automated Detection of Healthy and Infected Leaves in Agricultural Systems

Thu, 20 Jun 2024 00:00:00 +0000

Advances in computer vision and machine learning have transformed leaf disease detection by enabling efficient and accurate identification of subtle disease signs in leaves. Leveraging high-resolution imaging, pattern recognition algorithms, and deep learning models, researchers and farmers can now conduct automated detection across various plant species. The development focuses on sophisticated image processing techniques applied to diverse datasets captured under controlled conditions, ensuring comprehensive coverage of lighting, time, and weather variations. Expert annotation of infection stages and types enhances dataset reliability, while pre-processing stages such as resizing and normalization optimize image consistency for robust model training. Data augmentation techniques enrich dataset diversity, complemented by feature extraction methods like RGB color analysis, GLCM texture analysis, and shape descriptors to discern healthy and infected leaves with precision Validation through K-fold cross-validation ensures model reliability across diverse datasets, culminating in a deployable application for real-time leaf health monitoring. Results demonstrate significant advancements, with the proposed model achieving 92% accuracy, surpassing Logistic Regression (87%), Decision Tree (82%), and Support Vector Machine (79%). Over 10 epochs, the model achieves steady improvements to 95% training accuracy and 85% validation accuracy, underscoring its effectiveness. Implementing data augmentation boosts accuracy from 85% to 89%, while analysis of prediction errors refines model performance for enhanced automated plant health monitoring and precision agriculture applications. These advancements highlight the transformative impact of technology in safeguarding crop resilience and optimizing agricultural practices.

A Survey on Artificial Intelligence in Cybersecurity for Smart Agriculture: State-of-the-Art, Cyber Threats, Artificial Intelligence Applications, and Ethical Concerns

Sat, 20 Jul 2024 00:00:00 +0000

Wireless sensor networks and Internet of Things devices are revolutionizing the smart agriculture industry by increasing production, sustainability, and profitability as connectivity becomes increasingly ubiquitous. However, the industry has become a popular target for cyberattacks. This survey investigates the role of artificial intelligence (AI) in improving cybersecurity in smart agriculture (SA). The relevant literature for the study was gathered from Nature, Wiley Online Library, MDPI, ScienceDirect, Frontiers, IEEE Xplore Digital Library, IGI Global, Springer, Taylor & Francis, and Google Scholar. Of the 320 publications that fit the search criteria, 180 research papers were ultimately chosen for this investigation. The review described advancements from conventional agriculture to modern SA, including architecture and emerging technology. It digs into SA’s numerous uses, emphasizing its potential to transform farming efficiency, production, and sustainability. The growing reliance on SA introduces new cyber threats that endanger its integrity and dependability and provide a complete analysis of their possible consequences. Still, the research examined the essential role of AI in combating these threats, focusing on its applications in threat identification, risk management, and real-time response mechanisms. The survey also discusses ethical concerns such as data privacy, the requirement for high-quality information, and the complexities of AI implementation in SA. This study, therefore, intends to provide researchers and practitioners with insights into AI’s capabilities and future directions in the security of smart agricultural infrastructures. This study hopes to assist researchers, policymakers, and practitioners in harnessing AI for robust cybersecurity in SA, assuring a safe and sustainable agricultural future by comprehensively evaluating the existing environment and future trends.

Data Mining Utilizing Various Leveled Clustering Procedures on the Position of Workers in a Data Innovation Firm

Fri, 26 Jul 2024 00:00:00 +0000

The reason of this paper is to clarify dynamic clustering, the divisive and agglomerative dynamic clustering techniques. It fundamentally centers on the concept of the divisive different leveled shapes as well known as the top-down approach by creating a workflow appear, dendrograms, clustered data table which accumulated the clusters based the chosen property, and appear the isolated between each cluster with the assistance of an data mining device called Python. The DIANA dynamic approach utilized data tests of the list of laborers in a Data Advancement firm to induce clusters from the position column inside the data test table. In this work, we in addition executed genuine infers by creating barchart that shows up the ages of the chosen agent sets plotted against the positions which are the Engineers, Assistants, Workers and Troughs.

Introduction to Wi-Fi 7: A Review of History, Applications, Challenges, Economical Impact and Research Development

Mon, 26 Aug 2024 00:00:00 +0000

Wi-Fi 7, commonly referred to as IEEE 802.11be, is the most recent development in wireless communication technology. It provides significant improvements in terms of speed, capacity, and efficiency. The purpose of this study is to investigate Wi-Fi 7, a standard for wireless communication technology, with a particular focus on the technological advancements and security issues associated with it. In addition, it offers historical perspectives and investigates the institution's present capabilities as well as its potential for the future. The purpose of this study is to provide a complete examination of the development of Wi-Fi technology and its influence on a wide range of industries. This will allow researchers to look forward to and anticipate future advancements and breakthroughs. In addition, it made it possible to integrate Wi-Fi 7 in smart homes, intelligent transportation systems, healthcare services, and industrial automation by providing high-speed connectivity and lowering the amount of latency that occurs during interactions. Both positive and negative aspects, as well as future trends and developments, as well as economic consequences for a wide range of sectors, are investigated simultaneously. The findings of this study shed light on the possible uses of Wi-Fi 7 and its ability to transform a number of different industries, such as training and essential infrastructure. This review aims to enlighten developers and decision-makers about the capabilities and consequences of Wi-Fi 7. The objective is to provide a comprehensive understanding of the technology and promote its appropriate adoption in the current era of digital technology.

Enhancing Security and Performance in Vehicular Adhoc Networks: A Machine Learning Approach to Combat Adversarial Attacks

Mustafa Abdulfattah Habeeb , Yahya Layth Khaleel , Ahmed Raheem Abdulnabi — Fri, 20 Sep 2024 00:00:00 +0000

Integrating Machine Learning (ML) techniques into Vehicular Adhoc Networks (VANETs) provides promising features in autonomous driving and ITS applications. In this paper, DSRC data is used to evaluate the effectiveness of different ML models, including Naive Bayes, Random Forest, KNN, and Gradient Boosting, in normal and adversarial scenarios. Since the dataset is relatively imbalanced, the Synthetic Minority Over-sampling Technique (SMOTE) is employed for sampling, and defensive distillation for improving model resilience to adversarial perturbations. From the results, it is clear that models such as Gradient Boosting and Random Forest show high accuracy in both cases, thus showing the potential of using Machine Learning to improve VANET security and reliability when new threats appear. Through this research, the significance of the application of ML in the protection of vehicular communication in order to enhance both traffic safety and flow has been articulated.

Deep Learning Model for Hand Movement Rehabilitation

Fri, 11 Oct 2024 00:00:00 +0000

Electroencephalography (EEG) can control machines for human purposes, especially for disabled people doing rehabilitation exercises or regular tasks. Brain-computer interface (BCI) for Robotic hand uses deep learning to convert (EEG) brain activity into orders for robotic hand allowing users to move their hands right or left by the movement imagining. It could enable paralyzed individuals to perform basic hand movements and help in rehabilitation robots that help stroke patients regain hand function by offering guided exercises based on machine learning interpretations of their movements and intents. Artificial intelligence algorithms, particularly deep learning, classify and recognize patterns and intents implicit brainwaves as electroencephalography. However, EEG signals have a high degree of no stationarity, making their analysis challenging. As a result, selecting a suitable signal-processing strategy becomes critical. This study aimed to build a hybrid model to direct robotic arm movement, which applied movement direction and right or left classification. By integrating a pre-trained convolutional neural network (CNN) - the Inception V3 Model with a traditional machine learning algorithm (logistic regression (LR)), which is considered an extensive classification method, as well as identify a suitable signal processing method, the short-time Fourier transform (STFT) and continuous wavelet transform (CWT) to select the most accurate method for proposed model's classification. The training results of the proposed hybrid model show that STFT achieves higher average accuracy (0.998) than CWT (0.997), making it more precise for classifying the current dataset of nine subjects and enhancing the effectiveness of hybrid CNN model training. Similarly, the evaluation result of the average accuracy achieved by STFT is higher than that achieved by CWT in the evaluation metrics (0.997 > 0.797). This suggests that STFT is a superior choice for feature extraction, improving the generalization and robustness of the hybrid CNN model with logistic regression.

Segment Anything: A Review

Firas Hazzaa, Innocent Udoidiong, Akram Qashou, Sufian Yousef — Fri, 18 Oct 2024 00:00:00 +0000

Segment Anything (SA) is a state-of-the art method for universal object segmentation, which does not need task-specific training. Herein, we emphasize that SA can overcome the limitations of traditional segmentation frameworks based on requiring extensive manually annotated datasets and predefined architectures, as extensively documented in this review. SB supercharges performance and reduces cost by combining Mutual Information learning with an Efficient Transformer architecture, benefiting from a substantially larger pool of in-the-wild data. In this paper we review SA and its specific key innovations generality, resource boundedness, and scalability to large datasets. We also face obstacles such as data biases, computational complexity, real-world application issues and consider security as well as privacy in federated learning scenarios. It discusses areas for future research, such as increasing precision and robustness, incorporating the federated learning aspect and concerns regarding its ethical use in high risk domains of application. In this review, we highlight the transformative capacity that SA may bring to volume-wise object segmentation and urge the community to leverage on top of these new venues for a breakthrough in AI-vision systems.

Potato disease identification using transfer learning approaches

Tarza Hasan Abdullah — Sun, 03 Nov 2024 00:00:00 +0000

Potato crop is one of the prominent consumed foods by human beings. When potato crops are infected by diseases it affects farmers negatively and to run in a loss. Therefore, early detection of the potato crop disease can play a vital role in minimizing the loss of the farmers. Nowadays, artificial intelligence technologies, more specifically deep learning techniques, provide solutions to many crops disease-related problems. However, training deep learning models requires a high computational power and huge amount of data as they are data hungry models. Also, designing a custom CNN models a difficult task and there are some variations to be considered. To avoid these difficulties, we adopted two pretrained models of DenseNet121 and VGG19 through transfer learning approaches. The achieved accuracy for DenseNet121 and VGG19 models are 82.6% and 98.56% respectively. DenseNet121 model obtained the average precision, sensitivity, and F1-score of 88.19%, 82.53, and 82.04%, respectively. Whereas VGG19 yields 98.39% of precision, 98.39% of sensitivity, and 97.26% of F1-score in ternary 3-class classification (early blight vs healthy vs late blight).

Enhancing Motion Detection in Video Surveillance Systems Using the Three-Frame Difference Algorithm

Suhaib Qassem Yahya Al-Hashemi, Majid Salal Naghmash , Ahmad Ghandour — Fri, 15 Nov 2024 00:00:00 +0000

This paper outlines a methodology for motion detection in video surveillance systems, leveraging advanced algorithms and TCP/IP networks for real-time data acquisition and analysis. The primary focus is on the implementation of the Three-Frame Difference Algorithm, which detects moving targets by analyzing the differences between three consecutive video frames. This method significantly reduces redundant data transmission and storage, addressing the challenges posed by limited wireless network capabilities. The surveillance model, designed in MATLAB using SIMULINK, integrates computer vision systems with embedded coders to facilitate effective communication and processing of video data. The results demonstrate the system's capability to detect motion accurately under varying lighting conditions, thereby showcasing its potential applications in diverse fields such as security, robotics, and human-computer interaction.

Computer Networking and Cloud-Based Learning/Teaching Environment Using Virtual Labs Tools: A Review and Future Aspirations

Mishall Al-Zubaidie, Raad A. Muhajjar, Lauy Abdulwahid Shihabe — Sun, 24 Nov 2024 00:00:00 +0000

Physical laboratories for practical classrooms in computer networks can be prohibitively expensive, as well as requiring regular hardware/software upgrades. With Netkit and similar software, network laboratories can be set up in a computer lab, but the setup is complicated and each student still needs their own computer. Thanks to the Cloud-based infrastructure, Netkit is now available in pre-configured Amazon elastic compute Cloud (EC2) instances. This research reviews and introduces educational computer networks. Furthermore, a web interface is used to allow remote access to numerous lab scripts instantiated on the Cloud. This study also describes how to use the Cloud to set up and access Netkit, which is open-source and free, but only supports Linux computers. By hosting Netkit in the Cloud, where it can be accessed from any Internet-connected device regardless of the user’s device or operating system (OS), we have made Netkit portable. It was shown how the Cloud architecture of Amazon web services (AWS) holds NetKit-capable virtual machines that instructors and students can create on demand. This eliminates the need to set up specialised labs and enables students to gain practical experience with the principles presented in seminars on computer networks from any Internet-connected computer. Two other Cloud selling advantages are scalability and on-demand pricing. The horizontal elasticity of the Cloud service allows the system to be replicated to serve a much greater demand from users, so we need not worry about its physical configuration/structure. In addition, there is no need to worry about being charged for time not used. Finally, this research provides perspectives, obstacles and open issues that could present useful information for researchers in future studies.

Credit Card Fraud Detection based on Deep Learning Models

Thu, 05 Dec 2024 00:00:00 +0000

Credit card fraud detection (FD) protects consumers and financial institutions by identifying suspicious or unauthorized transactions. To improve security and reduce false positives, fraud detection systems can analyze transaction data patterns in real time using advanced machine learning (ML) and deep learning (DL). This paper exploits DL models to detects transactional data which includes anomalies through Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU) to verify data and mitigate fraud. The models used precision, recall, F1-score, and AUC on a balanced shared 559856-record Kaggle repository dataset. The RNN model detected anomalies with 99.39% accuracy, 0.9939 F1-score, and excellent recall. RNN shows promise as a real-time anomaly detection method with high performance and low computational cost.

An optimized model for Liver disease classification based on BPSO Using Machine learning models

Tue, 10 Dec 2024 00:00:00 +0000

Liver disease (LD) is a world health concern that requires accurate diagnostic methods. This study proposes an optimized machine learning model (ML) based on BPSO for LD classification using a shared public dataset from kaggle Indicates to liver patients from India. The paper used six ML models such as Random Forest (RF), Support Vector Machine (SVM), Dummy Classifier (DC), Extra Trees Classifier (ET), K-Nearest Neighbors (KNN), and Logistic Regression (LT) to evaluate the performance. Through observations we detected that ET achieved an accuracy of 79.82%. The BPSO hyperparameter optimization optimized ET to enhance accuracy to reach 85%. The paper used metrics such as accuracy, precision, recall, F1 score, and AUC. The results indicate that ML techniques with optimization have the potential to develop reliable diagnostic tools for LD.