Explore the vast range of titles available.

Coronaviruses (CoVs) are a large family of viruses that are common in many animal species, including camels, cattle, cats and bats. Animal CoVs, such as Middle East respiratory syndrome-CoV, severe acute respiratory syndrome (SARS)-CoV, and the new virus named SARS-CoV-2, rarely infect and spread among humans. On January 30, 2020, the International Health Regulations Emergency Committee of the World Health Organisation declared the outbreak of the resulting disease from this new CoV called ‘COVID-19’, as a ‘public health emergency of international concern’. This global pandemic has affected almost the whole planet and caused the death of more than 315,131 patients as of the date of this article. In this context, publishers, journals and researchers are urged to research different domains and stop the spread of this deadly virus. The increasing interest in developing artificial intelligence (AI) applications has addressed several medical problems. However, such applications remain insufficient given the high potential threat posed by this virus to global public health. This systematic review addresses automated AI applications based on data mining and machine learning (ML) algorithms for detecting and diagnosing COVID-19. We aimed to obtain an overview of this critical virus, address the limitations of utilising data mining and ML algorithms, and provide the health sector with the benefits of this technique. We used five databases, namely, IEEE Xplore, Web of Science, PubMed, ScienceDirect and Scopus and performed three sequences of search queries between 2010 and 2020. Accurate exclusion criteria and selection strategy were applied to screen the obtained 1305 articles. Only eight articles were fully evaluated and included in this review, and this number only emphasised the insufficiency of research in this important area. After analysing all included studies, the results were distributed following the year of publication and the commonly used data mining and ML algorithms. The results found in all papers were discussed to find the gaps in all reviewed papers. Characteristics, such as motivations, challenges, limitations, recommendations, case studies, and features and classes used, were analysed in detail. This study reviewed the state-of-the-art techniques for CoV prediction algorithms based on data mining and ML assessment. The reliability and acceptability of extracted information and datasets from implemented technologies in the literature were considered. Findings showed that researchers must proceed with insights they gain, focus on identifying solutions for CoV problems, and introduce new improvements. The growing emphasis on data mining and ML techniques in medical fields can provide the right environment for change and improvement.

Water quality monitoring plays a significant part in the transition towards intelligent and smart agriculture and provides an easy transition to automated monitoring of crucial components of human daily needs as new technologies are continuously developed and adopted in agricultural and human daily life (water). For the monitoring and management of water quality, this effort, however, requires reliable models with accurate and thorough datasets. Analyzing water quality monitoring models by utilizing sensors that gather water properties during live experiments is possible due to the necessity for precision in modeling. To convey numerous conclusions regarding the concerns, issues, difficulties, and research gaps that have existed throughout the past five years (2018–2022), this review article thoroughly examines the water quality literature. To find trustworthy peer-reviewed publications, several digital databases were searched and examined, including IEEE Xplore®, ScienceDirect, Scopus, and Web of Science. Only 50 articles out of the 946 papers obtained, were used in the study of the water quality monitoring research area. There are more rules for article inclusion in the second stage of the filtration process. Utilizing a real-time data acquisition system, the criteria for inclusion for the second phase of filtration looked at the implementation of water quality monitoring and characterization procedures. Reviews and experimental studies comprised most of the articles, which were divided into three categories. To organize the literature into articles with similar types of experimental conditions, a taxonomy of the three literature was created. Topics for recommendations are also provided to facilitate and speed up the pace of advancement in this field of study. By conducting a thorough analysis of the earlier suggested methodologies, research gaps are made clear. The investigation largely pointed out the problems in the accuracy of the models, the development of data-gathering systems, and the types of data used in the proposed frameworks. Finally, by examining critical topics required for the development of this research area, research directions toward smart water quality are presented.

Data scarcity is a major challenge when training deep learning (DL) models. DL demands a large amount of data to achieve exceptional performance. Unfortunately, many applications have small or inadequate data to train DL frameworks. Usually, manual labeling is needed to provide labeled data, which typically involves human annotators with a vast background of knowledge. This annotation process is costly, time-consuming, and error-prone. Usually, every DL framework is fed by a significant amount of labeled data to automatically learn representations. Ultimately, a larger amount of data would generate a better DL model and its performance is also application dependent. This issue is the main barrier for many applications dismissing the use of DL. Having sufficient data is the first step toward any successful and trustworthy DL application. This paper presents a holistic survey on state-of-the-art techniques to deal with training DL models to overcome three challenges including small, imbalanced datasets, and lack of generalization. This survey starts by listing the learning techniques. Next, the types of DL architectures are introduced. After that, state-of-the-art solutions to address the issue of lack of training data are listed, such as Transfer Learning (TL), Self-Supervised Learning (SSL), Generative Adversarial Networks (GANs), Model Architecture (MA), Physics-Informed Neural Network (PINN), and Deep Synthetic Minority Oversampling Technique (DeepSMOTE). Then, these solutions were followed by some related tips about data acquisition needed prior to training purposes, as well as recommendations for ensuring the trustworthiness of the training dataset. The survey ends with a list of applications that suffer from data scarcity, several alternatives are proposed in order to generate more data in each application including Electromagnetic Imaging (EMI), Civil Structural Health Monitoring, Medical imaging, Meteorology, Wireless Communications, Fluid Mechanics, Microelectromechanical system, and Cybersecurity. To the best of the authors’ knowledge, this is the first review that offers a comprehensive overview on strategies to tackle data scarcity in DL.

The Internet of Things (IoT) has been identified in various applications across different domains, such as in the healthcare sector. IoT has also been recognised for its revolution in reshaping modern healthcare with aspiring wide range prospects, including economical, technological and social. This study aims to establish IoT-based smart home security solutions for real-time health monitoring technologies in telemedicine architecture. A multilayer taxonomy is driven and conducted in this study. In the first layer, a comprehensive analysis on telemedicine, which focuses on the client and server sides, shows that other studies associated with IoT-based smart home applications have several limitations that remain unaddressed. Particularly, remote patient monitoring in healthcare applications presents various facilities and benefits by adopting IoT-based smart home technologies without compromising the security requirements and potentially large number of risks. An extensive search is conducted to identify articles that handle these issues, related applications are comprehensively reviewed and a coherent taxonomy for these articles is established. A total number of (n = 3064) are gathered between 2007 and 2017 for most reliable databases, such as ScienceDirect, Web of Science and Institute of Electrical and Electronic Engineer Xplore databases. Then, the articles based on IoT studies that are associated with telemedicine applications are filtered. Nine articles are selected and classified into two categories. The first category, which accounts for 22.22% (n = 2/9), includes surveys on telemedicine articles and their applications. The second category, which accounts for 77.78% (n = 7/9), includes articles on the client and server sides of telemedicine architecture. The collected studies reveal the essential requirement in constructing another taxonomy layer and review IoT-based smart home security studies. Therefore, IoT-based smart home security features are introduced and analysed in the second layer. The security of smart home design based on IoT applications is an aspect that represents a crucial matter for general occupants of smart homes, in which studies are required to provide a better solution with patient security, privacy protection and security of users’ entities from being stolen or compromised. Innovative technologies have dispersed limitations related to this matter. The existing gaps and trends in this area should be investigated to provide valuable visions for technical environments and researchers. Thus, 67 articles are obtained in the second layer of our taxonomy and are classified into six categories. In the first category, 25.37% (n = 17/67) of the articles focus on architecture design. In the second category, 17.91% (n = 12/67) includes security analysis articles that investigate the research status in the security area of IoT-based smart home applications. In the third category, 10.44% (n = 7/67) includes articles about security schemes. In the fourth category, 17.91% (n = 12/67) comprises security examination. In the fifth category, 13.43% (n = 9/67) analyses security protocols. In the final category, 14.92% (n = 10/67) analyses the security framework. Then, the identified basic characteristics of this emerging field are presented and provided in the following aspects. Open challenges experienced on the development of IoT-based smart home security are addressed to be adopted fully in telemedicine applications. Then, the requirements are provided to increase researcher’s interest in this study area. On this basis, a number of recommendations for different parties are described to provide insights on the next steps that should be considered to enhance the security of smart homes based on IoT. A map matching for both taxonomies is developed in this study to determine the novel risks and benefits of IoT-based smart home security for real-time remote health monitoring within client and server sides in telemedicine applications.

Musculoskeletal conditions affect an estimated 1.7 billion people worldwide, causing intense pain and disability. These conditions lead to 30 million emergency room visits yearly, and the numbers are only increasing. However, diagnosing musculoskeletal issues can be challenging, especially in emergencies where quick decisions are necessary. Deep learning (DL) has shown promise in various medical applications. However, previous methods had poor performance and a lack of transparency in detecting shoulder abnormalities on X-ray images due to a lack of training data and better representation of features. This often resulted in overfitting, poor generalisation, and potential bias in decision-making. To address these issues, a new trustworthy DL framework has been proposed to detect shoulder abnormalities (such as fractures, deformities, and arthritis) using X-ray images. The framework consists of two parts: same-domain transfer learning (TL) to mitigate imageNet mismatch and feature fusion to reduce error rates and improve trust in the final result. Same-domain TL involves training pre-trained models on a large number of labelled X-ray images from various body parts and fine-tuning them on the target dataset of shoulder X-ray images. Feature fusion combines the extracted features with seven DL models to train several ML classifiers. The proposed framework achieved an excellent accuracy rate of 99.2%, F1 Score of 99.2%, and Cohen’s kappa of 98.5%. Furthermore, the accuracy of the results was validated using three visualisation tools, including gradient-based class activation heat map (Grad CAM), activation visualisation, and locally interpretable model-independent explanations (LIME). The proposed framework outperformed previous DL methods and three orthopaedic surgeons invited to classify the test set, who obtained an average accuracy of 79.1%. The proposed framework has proven effective and robust, improving generalisation and increasing trust in the final results.

An intelligent remote prioritization for patients with high-risk multiple chronic diseases is proposed in this research, based on emotion and sensory measurements and multi-criteria decision making. The methodology comprises two phases: (1) a case study is discussed through the adoption of a multi-criteria decision matrix for high-risk level patients; (2) the technique for reorganizing opinion order to interval levels (TROOIL) is modified by combining it with an extended fuzzy-weighted zero-inconsistency (FWZIC) method over fractional orthotriple fuzzy sets to address objective weighting issues associated with the original TROOIL. In the first hierarchy level, chronic heart disease is identified as the most important criterion, followed by emotion-based criteria in the second. The third hierarchy level shows that Peaks is identified as the most important sensor-based criterion and chest pain as the most important emotion criterion. Low blood pressure disease is identified as the most important criterion for patient prioritization, with the most severe cases being prioritized. The results are evaluated using systematic ranking and sensitivity analysis.

The significance of deep learning techniques in relation to steady-state visually evoked potential- (SSVEP-) based brain-computer interface (BCI) applications is assessed through a systematic review. Three reliable databases, PubMed, ScienceDirect, and IEEE, were considered to gather relevant scientific and theoretical articles. Initially, 125 papers were found between 2010 and 2021 related to this integrated research field. After the filtering process, only 30 articles were identified and classified into five categories based on their type of deep learning methods. The first category, convolutional neural network (CNN), accounts for 70% (n=21/30). The second category, recurrent neural network (RNN), accounts for 10% (n=3/30). The third and fourth categories, deep neural network (DNN) and long short-term memory (LSTM), account for 6% (n=30). The fifth category, restricted Boltzmann machine (RBM), accounts for 3% (n=1/30). The literature’s findings in terms of the main aspects identified in existing applications of deep learning pattern recognition techniques in SSVEP-based BCI, such as feature extraction, classification, activation functions, validation methods, and achieved classification accuracies, are examined. A comprehensive mapping analysis was also conducted, which identified six categories. Current challenges of ensuring trustworthy deep learning in SSVEP-based BCI applications were discussed, and recommendations were provided to researchers and developers. The study critically reviews the current unsolved issues of SSVEP-based BCI applications in terms of development challenges based on deep learning techniques and selection challenges based on multicriteria decision-making (MCDM). A trust proposal solution is presented with three methodology phases for evaluating and benchmarking SSVEP-based BCI applications using fuzzy decision-making techniques. Valuable insights and recommendations for researchers and developers in the SSVEP-based BCI and deep learning are provided.

Evaluating remote sensing techniques (RST)-based water quality monitoring systems in hydrological applications is a complex multi-attribute decision-making (MADM) problem due to several integrated issues. These issues include the need to consider 15 evaluation criteria, the importance of these criteria in relation to their application and network infrastructure, data variation, and trade-offs and conflicts between them. Therefore, this study proposes an MADM integrated modelling approach to weight these criteria and rank the selected RSTs using the “Fuzzy Weighted with Zero Inconsistency” (FWZIC) method coupled with the “ Vlse-kriterijumska Optimizcija I Kaompromisno Resenje ” (VIKOR) method. First, a decision matrix was developed for the evaluation criteria and their intersection with various RSTs in different network categories. Next, the assessment criteria were weighted using FWZIC, followed by ranking the RSTs for each category using VIKOR. The findings reveal that the criteria weighting varied across categories, as assigned by specialists-based FWZIC. For example, the “data acquisition system complexity” criterion received the maximum weight ( w  = 0.081) for mesh-based sensing and the lowest ( w  = 0.057) for cellular-based sensing. Using the obtained weights and based on the ascending order of the performance score value ( Q ), various RSTs were benchmarked using VIKOR. This includes ( n  = 22) RSTs in fixed star, ( n  = 8) RSTs in mesh sensing, ( n  = 6) RSTs in cellular sensing, ( n  = 3) RSTs in fixed cable sensing, and ( n  = 2) RSTs in movable sensing. Finally, the evaluation process for the benchmarked RSTs involved systematic ranking, sensitivity analysis, and comparison analysis, which confirmed the robustness of the proposed approach across all RST categories.

Topical treatments with structural equation modelling (SEM) and an artificial neural network (ANN), including a wide range of concepts, benefits, challenges and anxieties, have emerged in various fields and are becoming increasingly important. Although SEM can determine relationships amongst unobserved constructs (i.e. independent, mediator, moderator, control and dependent variables), it is insufficient for providing non-compensatory relationships amongst constructs. In contrast with previous studies, a newly proposed methodology that involves a dual-stage analysis of SEM and ANN was performed to provide linear and non-compensatory relationships amongst constructs. Consequently, numerous distinct types of studies in diverse sectors have conducted hybrid SEM–ANN analysis. Accordingly, the current work supplements the academic literature with a systematic review that includes all major SEM–ANN techniques used in 11 industries published in the past 6 years. This study presents a state-of-the-art SEM–ANN classification taxonomy based on industries and compares the effort in various domains to that classification. To achieve this objective, we examined the Web of Science, ScienceDirect, Scopus and IEEE Xplore ® databases to retrieve 239 articles from 2016 to 2021. The obtained articles were filtered on the basis of inclusion criteria, and 60 studies were selected and classified under 11 categories. This multi-field systematic study uncovered new research possibilities, motivations, challenges, limitations and recommendations that must be addressed for the synergistic integration of multidisciplinary studies. It contributed two points of potential future work resulting from the developed taxonomy. First, the importance of the determinants of play, musical and art therapy adoption amongst autistic children within the healthcare sector is the most important consideration for future investigations. In this context, the second potential future work can use SEM–ANN to determine the barriers to adopting sensing-enhanced therapy amongst autistic children to satisfy the recommendations provided by the healthcare sector. The analysis indicates that the manufacturing and technology sectors have conducted the most number of investigations, whereas the construction and small- and medium-sized enterprise sectors have conducted the least. This study will provide a helpful reference to academics and practitioners by providing guidance and insightful knowledge for future studies.

When COVID-19 spread in China in December 2019, thousands of studies have focused on this pandemic. Each presents a unique perspective that reflects the pandemic’s main scientific disciplines. For example, social scientists are concerned with reducing the psychological impact on the human mental state especially during lockdown periods. Computer scientists focus on establishing fast and accurate computerized tools to assist in diagnosing, preventing, and recovering from the disease. Medical scientists and doctors, or the frontliners, are the main heroes who received, treated, and worked with the millions of cases at the expense of their own health. Some of them have continued to work even at the expense of their lives. All these studies enforce the multidisciplinary work where scientists from different academic disciplines (social, environmental, technological, etc.) join forces to produce research for beneficial outcomes during the crisis. One of the many branches is computer science along with its various technologies, including artificial intelligence, Internet of Things, big data, decision support systems (DSS), and many more. Among the most notable DSS utilization is those related to multicriterion decision making (MCDM), which is applied in various applications and across many contexts, including business, social, technological and medical. Owing to its importance in developing proper decision regimens and prevention strategies with precise judgment, it is deemed a noteworthy topic of extensive exploration, especially in the context of COVID-19-related medical applications. The present study is a comprehensive review of COVID-19-related medical case studies with MCDM using a systematic review protocol. PRISMA methodology is utilized to obtain a final set of ( n  = 35) articles from four major scientific databases (ScienceDirect, IEEE Xplore, Scopus, and Web of Science). The final set of articles is categorized into taxonomy comprising five groups: (1) diagnosis ( n  = 6), (2) safety ( n  = 11), (3) hospital ( n  = 8), (4) treatment ( n  = 4), and (5) review ( n  = 3). A bibliographic analysis is also presented on the basis of annual scientific production, country scientific production, co-occurrence, and co-authorship. A comprehensive discussion is also presented to discuss the main challenges, motivations, and recommendations in using MCDM research in COVID‐19-related medial case studies. Lastly, we identify critical research gaps with their corresponding solutions and detailed methodologies to serve as a guide for future directions. In conclusion, MCDM can be utilized in the medical field effectively to optimize the resources and make the best choices particularly during pandemics and natural disasters.