2692-4048ISSN (Online)
2770-0070ISSN (Print)
This study presents a comprehensive analysis of the existing techniques and applications of artificial intelligence (AI) to cardiovascular disease diagnosis. The application of AI to the diagnosis of cardiac diseases can enhance diagnostic precision, diagnostic throughput, and patient outcomes. This literature survey analyzes state-of-the-art AI-based methods, rates their efficiency, examines potential future research and development avenues, and finds challenges and limitations, providing a foundational overview of main developments in AI, machine learning, deep learning, and quantum computing in relation to heart disease prevention. This study seeks to guide the use of AI-based techniques for heart disease detection, having an ultimate objective of enhancing patient outcomes through research and development. This review mainly emphasizes the significance of further studying and advancing AI for its ability to revolutionize the diagnosis and management of heart diseases.
Read MoreDoi: https://doi.org/10.54216/FPA.110101
Vol. 11 Issue. 1 PP. 08-25, (2023)
The expression “COVID-19” has been the fiercest but most trending Google search since it first appeared in November 2019. Due to advances in mobile technology and sensors, Healthcare systems based on the Internet of Things are conceivable. Instead of the traditional reactive healthcare systems, these new healthcare systems can be proactive and preventive. This paper suggested a framework for real-time suspect detection based on the Internet of Things. In the early phases of predicting COVID-19, the framework evaluates the existence of the virus by extracting health variables obtained in real-time from sensors and other IoT devices, in order to better understand the behavior of the virus by collecting symptom data of COVID-19, In this paper, four machine learning models (Random Forest, Decision Tree, K-Nearest Neural Network, and Artificial Neural Network) are proposed, these data and applied as a machine learning model to obtain high diagnostic accuracy, however, it is noted that there is a problem when collecting clinical fusion data that is scarce and unbalanced, so a dataset augmented by Generative Adversarial Network (GAN) was used. Several algorithms achieved high levels of accuracy (ACC), including Random Forest (99%), and Decision Tree (99%), K-Nearest Neighbour (98%), and Artificial Neural Network (99%). These results show the ability of GANs to generate data and their ability to provide relevant data to efficiently manage Covid-19 and reduce the risk of its spread through accurate diagnosis of patients and informing health authorities of suspected cases.
Read MoreDoi: https://doi.org/10.54216/FPA.110102
Vol. 11 Issue. 1 PP. 26-36, (2023)
Data storage, software services, infrastructure services, and platform services are only some of the benefits of today's widespread use of cloud computing. Since most cloud services run via the internet, they are vulnerable to a comprehensive range of attacks that might end it the disclosure of sensitive information. The distributed denial-of-service (DDoS) is amongst the attacks that pose an active threat to the cloud environment and disrupts the provided services for the legitimate participants. The main aim of this review paper is to present the recent trends on sophisticated flooding attacks detection methods for cloud computing systems. The review only considers the papers published within the period of 2014 until 2022.This study aims to examine the various deep learning-based DDoS detection algorithms and machine learning used across different cloud environments. Also, the study covers the Sophisticated types of Flooding Attacks and the testing dataset. The review outcomes several research challenges, gaps and future research guidelines related to protection of DDoS attack in cloud computing environment.
Read MoreDoi: https://doi.org/10.54216/FPA.110103
Vol. 11 Issue. 1 PP. 37-56, (2023)
This paper investigates the process of selecting a hyperparameter for use in a kernel semiparametric regression model for fusion data, which is an important tool in various scientific study fields. The selection of the best model to use in advance is not a simple task, and one of the most fascinating current advances in the application is the use of hybrid metaheuristics algorithms to increase the exploration and exploitation capacity of traditional meta-heuristic algorithms. In this study, a hybrid optimization method that combines the pelican algorithm with the black hole algorithm is presented, which achieves a lower mean squared error (MSE) in comparison to other competing techniques. Data merging through the suggested hybrid metaheuristics algorithm gives superior performance in terms of computing time when compared to both the CV-method and the GCV-method. This work has practical implications for researchers and practitioners who use statistical modeling techniques in their work, especially those dealing with data merging for improved accuracy and efficiency.
Read MoreDoi: https://doi.org/10.54216/FPA.110104
Vol. 11 Issue. 1 PP. 57-69, (2023)
This paper presents an improved penalized regression-based clustering algorithm using a nature-inspired approach. Clustering is an unsupervised learning method widely used in data fusion mining, including gene analysis, to group unclassified fusion data based on their features. The proposed algorithm is an extension of the "Sum of Norms" model and aims to better estimate the data by fusing information from various sources. The performance of the proposed algorithm is evaluated on gene expression data. Results show that our approach outperforms other methods, indicating its potential impact on clustering research with data fusion.
Read MoreDoi: https://doi.org/10.54216/FPA.110105
Vol. 11 Issue. 1 PP. 70-76, (2023)
In autonomous vehicles, the control unit must be based on two main goals, first maintains the stability of the car second follows the desired path. All things considered, the controller's effectiveness is heavily dependent on the details of the steering system actuators. The necessary steering is set by a higher-order controller. The time delay of the steering actuator is one of the main features affecting the performance of the controller. While the artificial intelligence and artificial ethic are new apparatuses in autonomous vehicles but their ICs and electrical component are exposed to fusion. This paper primarily presents a more reliable system work during the fusion of multi-sensor information. We design the requirements of the steering system and the sureness of stability control in autonomous vehicles, also finding the suitable parameters for high-level control algorithms to compensate for time delay and ensure vehicle stability. The vehicle's steering angle response was obtained by testing the actuator of electric power steering (EPS) undergoing different speeds. In fact, using the identification of the system has been beneficial because obtaining the transfer function is easier than the actual methods which need the implementation of a mathematical model of the system. The system response of the Input-output has been defined via MATLAB. Full vehicle model simulation results indicate that the found adjustment parameter improves lane-tracking performance in a basic architecture by reducing oscillation and lateral error relative to other instances. The simplified steering system is the primary improvement brought by this effort.
Read MoreDoi: https://doi.org/10.54216/FPA.110106
Vol. 11 Issue. 1 PP. 77-86, (2023)
Every single day, thousands of crimes are perpetrated, and hundreds may be probably taking place right now throughout the world. Without a doubt, crime is viewed as a social blight. Nothing can truly stop it, no matter what is done. Surveillance cameras, on the other hand, can dramatically minimize it. Using public surveillance camera systems to prevent, document, and minimize crime can be a cost-effective solution. Installing enough cameras to detect crimes in progress and integrating technology to automate the monitoring of the live stream from these cameras will result in the most effective systems. Because of its self-learning characteristics, the advanced Artificial Intelligence surveillance system is constantly learning and improving. The Deep Learning Algorithms applied in this work processes videos using electronic devices like cameras in real-time termed as image processing, saving both human resources and a great deal of time. The highest accuracy of 86.6% was attained by Ensemble Model, followed by Inception Model with SGD Optimizer, Leaky Relu Activation Function giving an accuracy of 83.43%. Hence, anomalies were detected efficiently using decision making in real-time surveillance scenarios.
Read MoreDoi: https://doi.org/10.54216/FPA.110107
Vol. 11 Issue. 1 PP. 89-99, (2023)
With the use of multi-level features fusion, this work provides a new method for recognizing cognitive brain activity, which we term the Improved Multi-modal cognitive brain-imaging method (IMCBI). Identifying brain areas and basing judgments on insights into intelligent cognitive behavior for babies and adolescents presents a number of methodological issues that the suggested approach seeks to address. In order to understand how the brain functions during various motor, perceptual, and cognitive tasks, IMCBI employs smart methods for fusing data at several levels. This technique employs functional magnetic resonance imaging (fMRI) data to assess human behavioral activity in the brain while engaging in a variety of activities. It does so by combining an inter-subject retrieval strategy with deep neural networks (DNN). The research shows that the suggested method, which uses multi-level fusion of features, greatly raises the accuracy ratio to 95.63 percent, the sensitivity to 95.42 percent, and the specificity to 94.3 three point three percent. The findings demonstrate the method's efficacy in recognizing brain activity based on high-level cognitive ability, making it a useful tool for predicting clinical and behavioral responses.
Read MoreDoi: https://doi.org/10.54216/FPA.110108
Vol. 11 Issue. 1 PP. 100-113, (2023)
The free flow of people and products within metropolitan areas depends on well-managed transportation systems. However, public parking places in smart cities are often limited by traffic, causing cars and residents to waste time, money, and fuel. To counteract this issue, today's automobile systems combine information fusion with intelligent parking solutions. In this research, we present a Fuzzy Logic Integrated Machine Learning Algorithm (FL-MLA) for use in smart parking and traffic management in a metropolis. The FL-MLA use fuzzy induction to distinguish between parked and moving vehicles while calculating traffic flow. The suggested technique efficiently resolves the problem of locating suitable parking places by avoiding incorrect configurations that govern traffic management difficulties. Therefore, the FL-MLA is used in traffic management systems to boost performance metrics like efficiency ratio (98.1%) and accident detection (98.1%) based on simulation results like reduced energy consumption (95.3%), more accurate traffic estimation (97.9%), higher average daily park occupancy (97.2%), and higher efficiency ratio (98.1%).
Read MoreDoi: https://doi.org/10.54216/FPA.110109
Vol. 11 Issue. 1 PP. 114-128, (2023)