Optimizing AI Models for COVID-19 Detection and Forecasting: A Comprehensive Study

Abdelhameed

doi:https://doi.org/10.54216/MOR.010204

Journal of Intelligent Systems and Internet of Things | 2024

Neutrosophic Sentiment Analysis Method Using Orange Data Analysis

The present work tackles an urgent problem in the area of data analytics that is the shifting of sentiment against language in regards to human cognition. Although the science of data mining and machine learning has done much to address the problem of these tools, their scope is still limited regarding the management of human language which has inherent uncertainty and ambiguity. This research seeks to address this gap by illustrating how to apply a machine learning tool in a way that embraces the so-called uncertainty neutrality using the orange data analysis tool for analysis of visualized data. It is also important to note in the research that the combination of neutral and intelligent analysis with using applications such as orange increases the efficiency of sentiment classification and expands the theoretical scope of sentiment data analysis. Their findings underscore that this perspective seeks to illuminate details which other methods tend to ignore and hence offer a much more nuanced understanding of human cognition. Practically, this research presents an efficient paradigm as the new framework can be employed in market intelligence, evaluation of public policy and intelligent interface design, among others. As a result, this research does not only contribute to the body of knowledge within the profession of data science but also explores new dimensions in understanding human cognition.

groups

Janneth Ximena Iglesias Quintana mail -

Monica Alexandra Salame Ortiz mail -

Alipio Absalon Cadena Posso mail -

Joffre R. Paladines Rodriguez mail -

Bekbayeva Feruza Baxtiyerovna mail

link https://doi.org/10.54216/JISIoT.120114

Volume & Issue

Vol. Volume 12 / Iss. Issue 1

Details open_in_new

Metaheuristic Optimization Review | 2024

Leveraging Advanced Machine Learning for Pioneering Monkeypox Diagnosis: A New Paradigm in Infectious Disease Detection

Artificial intelligence (AI) is revolutionizing the problem solving of medical diagnosis, which has enduring issues, including early-stage disease, insufficient voluminous data, and diagnosis process ineffectiveness. This review demonstrates considerable progress in developing ML technologies, including monkeypox detection, Tuberculosis, and cancer diagnosis. CNNs have shown high efficiency in diagnostics; even InceptionV3, a transfer-learning model for clinicians, can reach 99.87% diagnostics. As privacy-preserving solutions, federated learning models work to improve diagnostic accuracy without increasing the exposure of individual data, and synthetic datasets derived from high-resolution techniques such as HiP-CT help deal with data scarcity by improving model construction and assessment. The hybrids of genome and metabolome integration helped enhance diagnostic accuracy measures, particularly for complex diseases like COVID-19, due to increased prognostic performance metrics using multiple biological information. However, few issues crop up even in modern society: Generalization of the model is an issue due to a lack of data, especially for rare conditions, and increased computational power requirements for most ML models pose a problem for implementation in low-resource environments. Prominent ethical issues incorporating algorithm prejudices and the ‘black box’ concept spotlight the requisite of an explainable AI (XAI) framework to provide visibility and credence in the medical facility. Possible directions in development, such as the standardization of frameworks, enhancing computational support, and integration of different fields, provide ways to address these challenges. When tackled, these challenges create the possibility of revamping global healthcare through suitable and scalable approaches informed by ML technologies that align with the patient’s needs, leading to better practices and, consequently, better health.

groups

El-Sayed M. El-kenawy mail

link https://doi.org/10.54216/MOR.010101

Volume & Issue

Vol. Volume 1 / Iss. Issue 1

Details open_in_new

Metaheuristic Optimization Review | 2024

Stem Cells and Regenerative Medicine: A Review of Artificial Intelligence Techniques for Stem Cell Differentiation

Following this background, this review discusses the advanced technologies such as AI, micro-fluids, and automated platforms that this differentiation protocol could help achieve in regenerative medicine. Stem cell research, essential in tissue engineering, disease modeling, and drug development, is challenging through heterogeneity, scalability, and reproducibility, as observed in differentiation procedures. Machine learning and deep learning patterns have become more effective in predicting cellular behavior, tracking cellular stations, and optimizing differentiation methods for current stem cell technology. These methods also reduce observer bias, increase the throughput of high-throughput screening, and advance modeling to precise therapeutic applications. At the same time, microfluidic and automated systems provide nearly perfect control over differentiation stimuli, recreating the in vivo conditions with the ability to control spatial and temporal gradients. This integration between AI and microengineering has promoted 3D culture systems, lab-on-a-chip technologies, and biosensors, enabling reproducible and efficient differentiation results. There is still much to accomplish, such as the problem of obtaining uniform stem cell populations or decoding the tissue context. This field incorporates several interdisciplinary advancements such as stimuli-responsive systems and computational modeling; it envisages new horizons in regenerative medicine, transforming stem cell-based therapeutic technologies to their optimum level for personalized medicine and other advanced tissue engineering applications.

groups

Nima Khodadadi mail -

Benyamin Abdollahzadeh mail

link https://doi.org/10.54216/MOR.010102

Volume & Issue

Vol. Volume 1 / Iss. Issue 1

Details open_in_new

Metaheuristic Optimization Review | 2024

Improving Tuberculosis Diagnosis and Forecasting Through Machine Learning Techniques: A Systematic Review

Tuberculosis (TB) is ranked as one of the leading causes of death from infectious diseases in the present world, causing important health and economic consequences in the different developing countries. The practices of traditional diagnostic approaches, although still expected, are associated with relativity, slowness, and organs, besides being confined to visual observations and touch. The new and increased capacity in advanced machine learning is a promising area that has shown potential in improving the diagnosis of TB, as well as identifying drug resistance and disease management. This review presents various aspects of using ML in diagnosing and managing TB disease based on its various categories of models, including deep learning, hybrid approach, and the metabolomics approach. Some of these methods have been very effective, with high diagnostic performance improvements in sensitivity, specificity and accuracy; Furthermore, ML has been used to analyze the molecular picture of TB and to find drug targets of the disease toward future targeted therapies. As seen with the integration of ML, substantial benefits are provided by the solutions proposed. However, questions concerning the quality of data, interpretations of ML models and ethical problems hinder further application. This review concludes with the idea that ML can transform the diagnosis and management of TB and calls for more investment in developing this field to overcome these barriers to global health.

groups

Abdelaziz Rabehi mail -

Pushan Kumar Dutta mail

link https://doi.org/10.54216/MOR.010103

Volume & Issue

Vol. Volume 1 / Iss. Issue 1

Details open_in_new

Metaheuristic Optimization Review | 2024

Innovations in Machine Learning Models for Hepatitis Diagnosis and Disease Progression Prediction

Chronic liver disease (CLD) is a group of conditions for which up to half of the global population remains at risk and causes serious complications: liver cirrhosis and liver cancer. Therefore, early diagnosis and proper treatment of these diseases enhance the prognosis of patients suffering from CLD. This review paper explores how machine learning (ML) techniques are used in practice to diagnose, prognosis, and treat chronic liver diseases. Continuing with more specific examples of collected data from the results of several studies, their more comprehensive implementation is expected to improve the respective management processes and the detection of liver disease in patients more accurately. The review further discusses the various ML methods, including supervised and unsupervised learning, neural network, and ensemble learning, also applied to the estimation of risk felt by the patients, suggesting a course of treatment or how far the disease has progressed. While the inclusion of ML technology in the field of Hepatology is progressing well, some issues like model diversity, applicability of models, and concerns about ethics still pose challenges. This paper points out the importance of working in teams from various fields to develop appropriate mechanisms for dealing with these issues and adequately use ML for clinics. In conclusion, the results indicate that there is a possibility that ML will change the management of chronic liver diseases, which in turn will lead to the development of innovative treatment methods and better patient management.

groups

Marwa M. Eid mail -

Wei Hong Lim mail

link https://doi.org/10.54216/MOR.010104

Volume & Issue

Vol. Volume 1 / Iss. Issue 1

Details open_in_new

Metaheuristic Optimization Review | 2024

Machine Learning Approaches for Malaria Risk Prediction and Detection: Trends and Insights

The current review summarizes the latest trends in malaria literature, emphasizing transmission ecology, new diagnostics and treatment. It stresses the additional focus on the transmission, according to the spatiotemporal models and predictive analytics, which help identify periods and the locations with the most significant risk, noting that these processes should consider the environmental factors. The change in the diagnostic approach, especially the introduction of artificial intelligence techniques such as deep learning, has improved the rate and precision at which malaria parasites are diagnosed in resource-limited countries where time is of the essence. Furthermore, there have been significant advances in drug discovery due to machine learning applications that have made it quicker to find new antimalarial drugs in the face of drug resistance. Despite these developments, there are still problems such as drug resistance, socio-economic disparities, and the environment that are being altered and still require an integrated and transdisciplinary approach. Combining these determinants is indispensable for eliminating these challenges and further promoting global efforts to control malaria.

groups

Ahmed Mohamed Zaki mail -

Khaled Sh. Gaber mail -

Faris H. Rizk mail -

Mahmoud Elshabrawy Mohamed mail

link https://doi.org/10.54216/MOR.010105

Volume & Issue

Vol. Volume 1 / Iss. Issue 1

Details open_in_new

Metaheuristic Optimization Review | 2024

Machine Learning in Public Health Forecasting and Monitoring the Zika Virus

The Zika virus is a severe public health threat all across the world, owing to its spreading mechanism through Aedes mosquitoes and its ability to result in extreme neurological diseases, which include the congenital Zika syndrome and the Guillain-Barré syndrome, amongst others. Conventional monitoring techniques often fail because many asymptomatic cases render early diagnosis challenging. Machine learning (ML) techniques can be seen as a constructive development in addressing this challenge, which entails predicting and tracking the spread of diseases such as Zika through extensive and complex datasets. Data analytic ML systems also enhance early warning systems and situational uplift by using data from social media, climate history, and genetics. This helps reasonably to predict the mosquito population biologically and the environmental factors that favor the spread of the virus for a more practical approach from the public health sector. Over and above, some issues are still pending, especially regarding the quality of data, understanding the models and how to apply such models within the current health systems. These factors must be solved to implement ML successfully in surveillance practice. This review provides an overview of the issue, stating the potential of machine learning applications in the development of public health, whose actions focus on Zika and other diseases transmitted by vectors.

groups

El-Sayed M. El-kenawy mail -

Marwa M. Eid mail -

Laith Abualigah mail

link https://doi.org/10.54216/MOR.010201

Volume & Issue

Vol. Volume 1 / Iss. Issue 2

Details open_in_new

Metaheuristic Optimization Review | 2024

Leveraging Machine Learning Algorithms for Early Detection and Prediction of Dengue Fever

This paper explores the potential of machine learning (ML) in revolutionizing the screening and prognosis of dengue fever, a pervasive viral illness transmitted through the bite of infected mosquitoes prevalent in tropical and subtropical areas. Historically, traditional approaches to monitoring outbreaks have been hampered by a lack of precision and timing, creating an opportunity for machine learning to rectify datasets and uncover patterns that enhance accuracy. The paper introduces Random Forests, Support Vector Machines, Neural Networks, and combined classification models, along with their advantages, disadvantages, and the potential for incorporating external data such as climatic factors, population data, real-time Twitter data, etc. The results demonstrate significant increases in accuracy from the models, but it is clear that their applicability is contingent on localized approaches suitable for the regions. This underscores the importance of the quality and completeness of data used in the models. Current research indicates that data availability and the refinement of these models require a collective approach. The work underscores the potential of ML to redefine the paradigm of outbreak management in dengue and other vector-borne diseases, offering hope for improved public health worldwide.

groups

Marwa M. Eid mail -

Christos Gatzoulis mail -

Osama Al Abedallat mail

link https://doi.org/10.54216/MOR.010202

Volume & Issue

Vol. Volume 1 / Iss. Issue 2

Details open_in_new

Metaheuristic Optimization Review | 2024

A Review of Machine Learning Techniques for Early Detection of Alzheimer's disease

This review aims to discuss the use of AI and ML in diagnosing and managing neurodegenerative diseases, with particular emphasis on AD and MCI. Emerging innovations present in depth the effectiveness of using ML models such as SVM, random forests, CNNs, and new frameworks such as quantum-classical neural networks on data obtained from MRI imaging, EEG signals, genetic makers and sociodemographic data. Widely used research findings demonstrate that these tools offer seemingly higher detection rates, sensitivity, and specificity than traditional diagnostic techniques for identifying and diagnosing early-stage illnesses. Some of them are techniques based on analyzing EEG time-frequency bands, combining MRI and PET data integration approaches, and creating telemedicine services to overcome geographical barriers. Furthermore, interpretable AI models improve clinical relevance by providing decision and trust support among practitioners. While these achievements are notable, the following limitations need to be noted, thus making it easier to establish the generalizability of the results and ways of using datasets that are free from bias and difficulties associated with applying AI in clinical settings. There are pressing questions regarding patients' rights and privacy, the issue of homogenization and standardization of data, and the distribution and accessibility of AI tools across industries as well as within the same region. More studies should be conducted to expand AI applications, use a more diverse dataset, and promote cooperation between representatives of various fields of science to ensure that technological advancement meets clinical demands. It also includes new methods like Vision Transformers and Quantum Computing Enhanced Deep Learning to overcome diagnostic issues in time-consuming and multi-parametric data analysis. These gaps can be closed with the help of AI and ML to enhance diagnostic accuracy, select the right treatment strategy, and risk assessment for the long-term management of NDs. In conclusion, this review similarly reaffirms how stunning AI's role is in improving future neurodegenerative disease care. For this reason, the deployment process must be done sensibly to enhance the patient's value most appropriately.

groups

El-Sayed M. El-kenawy mail

link https://doi.org/10.54216/MOR.010203

Volume & Issue

Vol. Volume 1 / Iss. Issue 2

Details open_in_new

Metaheuristic Optimization Review | 2024

Optimizing AI Models for COVID-19 Detection and Forecasting: A Comprehensive Study

This systematic review explores the use of artificial intelligence (AI) and machine learning (ML) during the COVID-19 disease outbreak. AI/ML models may interpret medical images, auditory input, and patient records to diagnose early enough, thus enhancing the likelihood of positive patient outcomes. Coupled with optimization algorithms, deep learning methods have predicted COVID-19 from chest X-rays and CT scans with unprecedented high accuracy. This review, therefore, synthesizes the existing literature and looks at the significant emphases, gaps, and potential trends of applying AI in diagnosing COVID-19 and forecasting outbreaks. Further, the advancement of AI and ML in this domain needs to be known to enhance global preventive diagnostic techniques for future pandemics.

groups

Abdelhameed Ibrahim mail

link https://doi.org/10.54216/MOR.010204

Volume & Issue

Vol. Volume 1 / Iss. Issue 2

Details open_in_new