Pub Date : 2025-02-01DOI: 10.1016/j.cosrev.2025.100730
Khosro Rezaee
Autism Spectrum Disorder (ASD) is a multifaceted neurodevelopmental condition characterized by social communication challenges, repetitive behaviors, and restricted interests. Early and accurate diagnosis is paramount for effective intervention and treatment, significantly improving the quality of life for individuals with ASD. This comprehensive review aims to elucidate the various methodologies employed in the automated diagnosis of ASD, providing a comparative analysis of their diagnostic accuracy, privacy considerations, non-invasiveness, cost implications, computational complexity, and feasibility for clinical and therapeutic use. The study encompasses a wide range of techniques including neuroimaging, EEG signal analysis, speech and crying signal analysis, eye tracking, facial recognition, and body movement analysis, highlighting their potential and limitations in the context of ASD diagnosis. By exploring these diverse diagnostic approaches, the review seeks to offer insights into the most promising methods and identify areas for future research and development.
{"title":"Machine learning in automated diagnosis of autism spectrum disorder: a comprehensive review","authors":"Khosro Rezaee","doi":"10.1016/j.cosrev.2025.100730","DOIUrl":"https://doi.org/10.1016/j.cosrev.2025.100730","url":null,"abstract":"Autism Spectrum Disorder (ASD) is a multifaceted neurodevelopmental condition characterized by social communication challenges, repetitive behaviors, and restricted interests. Early and accurate diagnosis is paramount for effective intervention and treatment, significantly improving the quality of life for individuals with ASD. This comprehensive review aims to elucidate the various methodologies employed in the automated diagnosis of ASD, providing a comparative analysis of their diagnostic accuracy, privacy considerations, non-invasiveness, cost implications, computational complexity, and feasibility for clinical and therapeutic use. The study encompasses a wide range of techniques including neuroimaging, EEG signal analysis, speech and crying signal analysis, eye tracking, facial recognition, and body movement analysis, highlighting their potential and limitations in the context of ASD diagnosis. By exploring these diverse diagnostic approaches, the review seeks to offer insights into the most promising methods and identify areas for future research and development.","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"168 1","pages":""},"PeriodicalIF":12.9,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143072454","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-30DOI: 10.1016/j.cosrev.2025.100728
Gaetano Perrone, Simon Pietro Romano
WebAssembly is revolutionizing the approach to developing modern applications. Although this technology was born to create portable and performant modules in web browsers, currently, its capabilities are extensively exploited in multiple and heterogeneous use-case scenarios. With the extensive effort of the community, new toolkits make the use of this technology more suitable for real-world applications. In this context, it is crucial to study the liaisons between the WebAssembly ecosystem and software security. Indeed, WebAssembly can be a medium for improving the security of a system, but it can also be exploited to evade detection systems or for performing crypto-mining activities. In addition, programs developed in low-level languages such as C can be compiled in WebAssembly binaries, and it is interesting to evaluate the security impacts of executing programs vulnerable to attacks against memory in the WebAssembly sandboxed environment. Also, WebAssembly has been designed to provide a secure and isolated environment, but such capabilities should be assessed in order to analyze their weaknesses and propose new mechanisms for addressing them. Although some research works have provided surveys of the most relevant solutions aimed at discovering WebAssembly vulnerabilities or detecting attacks, at the time of writing there is no comprehensive review of security-related literature in the WebAssembly ecosystem. We aim to fill this gap by proposing a comprehensive review of research works dealing with security in WebAssembly. We analyze 147 papers by identifying seven different security categories.
{"title":"WebAssembly and security: A review","authors":"Gaetano Perrone, Simon Pietro Romano","doi":"10.1016/j.cosrev.2025.100728","DOIUrl":"https://doi.org/10.1016/j.cosrev.2025.100728","url":null,"abstract":"WebAssembly is revolutionizing the approach to developing modern applications. Although this technology was born to create portable and performant modules in web browsers, currently, its capabilities are extensively exploited in multiple and heterogeneous use-case scenarios. With the extensive effort of the community, new toolkits make the use of this technology more suitable for real-world applications. In this context, it is crucial to study the liaisons between the WebAssembly ecosystem and software security. Indeed, WebAssembly can be a medium for improving the security of a system, but it can also be exploited to evade detection systems or for performing crypto-mining activities. In addition, programs developed in low-level languages such as C can be compiled in WebAssembly binaries, and it is interesting to evaluate the security impacts of executing programs vulnerable to attacks against memory in the WebAssembly sandboxed environment. Also, WebAssembly has been designed to provide a secure and isolated environment, but such capabilities should be assessed in order to analyze their weaknesses and propose new mechanisms for addressing them. Although some research works have provided surveys of the most relevant solutions aimed at discovering WebAssembly vulnerabilities or detecting attacks, at the time of writing there is no comprehensive review of security-related literature in the WebAssembly ecosystem. We aim to fill this gap by proposing a comprehensive review of research works dealing with security in WebAssembly. We analyze 147 papers by identifying seven different security categories.","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"13 1","pages":""},"PeriodicalIF":12.9,"publicationDate":"2025-01-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143072455","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-22DOI: 10.1016/j.cosrev.2025.100729
Adamu Tafida, Wesam Salah Alaloul, Noor Amila Bt Wan Zawawi, Muhammad Ali Musarat, Adamu Abubakar Sani
Road infrastructure networks are crucial in facilitating smart mobility, as indicated by the emergence of innovative transportation concepts that offer improved efficiency and environmental sustainability. This study seeks to review the literature regarding road pavement condition assessment performance improvement tools which utilize various computer vision and photogrammetry tools aided by machine learning algorithms towards mitigating challenges encountered and promoting smart transportation trends. A comprehensive search of available literature was conducted, and relevant studies were analyzed to identify computer vision and photogrammetry tools used, learning-based algorithms deployed and contribution to the improvement of road infrastructure to aid smart transportation. The review considered emerging challenges of the techniques, identified research gaps and explored the potentials of the techniques as it relates to aiding wider acceptance of the implementation of autonomous vehicles and smart transportation The study found gaps in knowledge relating to the computer vision (CV) and photogrammetry tools standardization of evaluation parameters, the applicability of the models for real-time assessment and implications regarding the adoption of autonomous vehicles and smart transportation which were not sufficiently considered in the previous cited literature. Future research areas were highlighted and its implication regarding the promotion of smart transportation.
{"title":"Advancing smart transportation: A review of computer vision and photogrammetry in learning-based dimensional road pavement defect detection","authors":"Adamu Tafida, Wesam Salah Alaloul, Noor Amila Bt Wan Zawawi, Muhammad Ali Musarat, Adamu Abubakar Sani","doi":"10.1016/j.cosrev.2025.100729","DOIUrl":"https://doi.org/10.1016/j.cosrev.2025.100729","url":null,"abstract":"Road infrastructure networks are crucial in facilitating smart mobility, as indicated by the emergence of innovative transportation concepts that offer improved efficiency and environmental sustainability. This study seeks to review the literature regarding road pavement condition assessment performance improvement tools which utilize various computer vision and photogrammetry tools aided by machine learning algorithms towards mitigating challenges encountered and promoting smart transportation trends. A comprehensive search of available literature was conducted, and relevant studies were analyzed to identify computer vision and photogrammetry tools used, learning-based algorithms deployed and contribution to the improvement of road infrastructure to aid smart transportation. The review considered emerging challenges of the techniques, identified research gaps and explored the potentials of the techniques as it relates to aiding wider acceptance of the implementation of autonomous vehicles and smart transportation The study found gaps in knowledge relating to the computer vision (CV) and photogrammetry tools standardization of evaluation parameters, the applicability of the models for real-time assessment and implications regarding the adoption of autonomous vehicles and smart transportation which were not sufficiently considered in the previous cited literature. Future research areas were highlighted and its implication regarding the promotion of smart transportation.","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"109 1","pages":""},"PeriodicalIF":12.9,"publicationDate":"2025-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143027368","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The Artificial Hummingbird Algorithm (AHA) is a metaheuristic optimization technique inspired by the behaviours and foraging strategies of hummingbirds. Known for their extraordinary agility and accuracy in collecting nectar, hummingbirds provide an exemplary framework for tackling complex optimization problems. Developed by Zhao et al. in 2022, AHA has swiftly attracted interest within the research community because to its exceptional performance and adaptability. This study provides a detailed and comprehensive review of AHA, exploring the diverse versions and modifications published in multiple research papers since its inception in 2022, with 23 % appearing in international conference papers and 75 % in esteemed peer-reviewed journals. The variants of AHA covered in this paper include 55 % of classical AHA, 17 % of improved AHA, 11 % of hybridization, 2 % of binary, 15 % of multi-objective variants, respectively. Furthermore, the applications of AHA illustrate its effectiveness and adaptability across various fields, with 42 % in power and control engineering, 11 % in optimizing deep learning models, 10 % in engineering design challenges, and 8 % in renewable energy sources. The algorithm has been utilized substantially in the domain of IoT, wireless sensor networks, wind energy, and fog computing. Furthermore, we also evaluate the performance of the AHA in the image clustering domain, and the findings revealed that the AHA performs better in comparison to the other tested methods. The main objectives of this study are to deliver a comprehensive review of AHA, emphasizing its novel methodology, and analyzing its various variants and their applications in numerous fields. As nature-inspired optimization methods continue to evolve, this survey paper expected to serves as a valuable resource for researchers aiming to gain a comprehensive understanding of AHA, its progression, and its diverse applications in solving complex optimization problems.
{"title":"Artificial hummingbird algorithm: Theory, variants, analysis, applications, and performance evaluation","authors":"Buddhadev Sasmal, Arunita Das, Krishna Gopal Dhal, Ramesh Saha, Rebika Rai, Totan Bharasa, Essam H. Houssein","doi":"10.1016/j.cosrev.2025.100727","DOIUrl":"https://doi.org/10.1016/j.cosrev.2025.100727","url":null,"abstract":"The Artificial Hummingbird Algorithm (AHA) is a metaheuristic optimization technique inspired by the behaviours and foraging strategies of hummingbirds. Known for their extraordinary agility and accuracy in collecting nectar, hummingbirds provide an exemplary framework for tackling complex optimization problems. Developed by Zhao et al. in 2022, AHA has swiftly attracted interest within the research community because to its exceptional performance and adaptability. This study provides a detailed and comprehensive review of AHA, exploring the diverse versions and modifications published in multiple research papers since its inception in 2022, with 23 % appearing in international conference papers and 75 % in esteemed peer-reviewed journals. The variants of AHA covered in this paper include 55 % of classical AHA, 17 % of improved AHA, 11 % of hybridization, 2 % of binary, 15 % of multi-objective variants, respectively. Furthermore, the applications of AHA illustrate its effectiveness and adaptability across various fields, with 42 % in power and control engineering, 11 % in optimizing deep learning models, 10 % in engineering design challenges, and 8 % in renewable energy sources. The algorithm has been utilized substantially in the domain of IoT, wireless sensor networks, wind energy, and fog computing. Furthermore, we also evaluate the performance of the AHA in the image clustering domain, and the findings revealed that the AHA performs better in comparison to the other tested methods. The main objectives of this study are to deliver a comprehensive review of AHA, emphasizing its novel methodology, and analyzing its various variants and their applications in numerous fields. As nature-inspired optimization methods continue to evolve, this survey paper expected to serves as a valuable resource for researchers aiming to gain a comprehensive understanding of AHA, its progression, and its diverse applications in solving complex optimization problems.","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"87 1","pages":""},"PeriodicalIF":12.9,"publicationDate":"2025-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143027371","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-17DOI: 10.1016/j.cosrev.2025.100726
Muhammad Harith Noor Azam, Farida Ridzuan, M. Norazizi Sham Mohd Sayuti, A H Azni, Nur Hafiza Zakaria, Vidyasagar Potdar
Cover selection is the process of selecting a suitable cover for steganography. Cover selection is crucial to maintain the steganographic characteristics performances and further avoid detection of hidden messages by eavesdroppers. Numerous existing reviews have focused mainly on the implementation and performance of steganography methods. Existing reviews have demonstrated inadequate depth of analysis and a lack of the number of articles reviewed. Thus, this article systematically reviews 34 cover selection methods for steganography in five databases including Web of Science, IEEE Xplore, ScienceDirect, Scopus, and Springer. The results include a trend analysis concerning existing cover selection algorithms for steganography. This article also establishes four novel classifications for cover selection methods. Recommendations on the implementation and design for cover selection method based on each class are provided. Analysis of the elements including cover types, datasets, searching methods, evaluation metrics for searching methods, cover selection attributes and its performance evaluations are also provided. An in-depth discussion on how cover types, searching method and evaluation metrics for searching method affects the steganography characteristics are also presented. This review offers valuable insights for researchers in developing new methods and enhance steganography systems for secure data communication.
封面选择是为隐写术选择合适的封面的过程。掩体选择对于保持隐写特征性能和进一步避免被窃听者发现隐藏信息至关重要。许多现有的评论主要集中在隐写方法的实现和性能上。现有的综述表明，分析的深度不足，而且综述的文章数量不足。因此，本文系统地综述了Web of Science、IEEE explore、ScienceDirect、Scopus和b施普林格等5个数据库中34种隐写术的封面选择方法。结果包括对隐写术现有封面选择算法的趋势分析。本文还建立了四种新的封面选择方法分类。提出了基于各类别的封面选择方法的实施和设计建议。分析了覆盖类型、数据集、搜索方法、搜索方法的评价指标、覆盖选择属性及其性能评价等要素。深入讨论了覆盖类型、搜索方法和搜索方法的评价指标对隐写特性的影响。这一综述为研究人员开发新的隐写方法和增强安全数据通信的隐写系统提供了有价值的见解。
{"title":"A systematic review on cover selection methods for steganography: Trend analysis, novel classification and analysis of the elements","authors":"Muhammad Harith Noor Azam, Farida Ridzuan, M. Norazizi Sham Mohd Sayuti, A H Azni, Nur Hafiza Zakaria, Vidyasagar Potdar","doi":"10.1016/j.cosrev.2025.100726","DOIUrl":"https://doi.org/10.1016/j.cosrev.2025.100726","url":null,"abstract":"Cover selection is the process of selecting a suitable cover for steganography. Cover selection is crucial to maintain the steganographic characteristics performances and further avoid detection of hidden messages by eavesdroppers. Numerous existing reviews have focused mainly on the implementation and performance of steganography methods. Existing reviews have demonstrated inadequate depth of analysis and a lack of the number of articles reviewed. Thus, this article systematically reviews 34 cover selection methods for steganography in five databases including Web of Science, IEEE Xplore, ScienceDirect, Scopus, and Springer. The results include a trend analysis concerning existing cover selection algorithms for steganography. This article also establishes four novel classifications for cover selection methods. Recommendations on the implementation and design for cover selection method based on each class are provided. Analysis of the elements including cover types, datasets, searching methods, evaluation metrics for searching methods, cover selection attributes and its performance evaluations are also provided. An in-depth discussion on how cover types, searching method and evaluation metrics for searching method affects the steganography characteristics are also presented. This review offers valuable insights for researchers in developing new methods and enhance steganography systems for secure data communication.","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"55 1","pages":""},"PeriodicalIF":12.9,"publicationDate":"2025-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142988079","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Medical image segmentation plays an important role in computer-aided diagnosis. Attention mechanisms that distinguish important parts from irrelevant parts have been widely used in medical image segmentation tasks. This paper systematically reviews the basic principles of attention mechanisms and their applications in medical image segmentation. First, we review the basic concepts of attention mechanism and formulation. Second, we surveyed about 200 articles related to medical image segmentation, and divided them into three groups based on their attention mechanisms, Pre-Transformer attention, Transformer attention and Mamba-related attention. In each group, we deeply analyze the attention mechanisms from three aspects based on the current literature work, i.e., the principle of the mechanism (what to use), implementation methods (how to use), and application tasks (where to use). We also thoroughly analyzed the advantages and limitations of their applications to different tasks. Finally, we summarize the current state of research and shortcomings in the field, and discuss the potential challenges in the future, including task specificity, robustness, standard evaluation, etc. We hope that this review can showcase the overall research context of traditional, Transformer and Mamba attention methods, provide a clear reference for subsequent research, and inspire more advanced attention research, not only in medical image segmentation, but also in other image analysis scenarios. Finally, we maintain the paper list and open-source code at here.
{"title":"Advances in attention mechanisms for medical image segmentation","authors":"Jianpeng Zhang, Xiaomin Chen, Bing Yang, Qingbiao Guan, Qi Chen, Jian Chen, Qi Wu, Yutong Xie, Yong Xia","doi":"10.1016/j.cosrev.2024.100721","DOIUrl":"https://doi.org/10.1016/j.cosrev.2024.100721","url":null,"abstract":"Medical image segmentation plays an important role in computer-aided diagnosis. Attention mechanisms that distinguish important parts from irrelevant parts have been widely used in medical image segmentation tasks. This paper systematically reviews the basic principles of attention mechanisms and their applications in medical image segmentation. First, we review the basic concepts of attention mechanism and formulation. Second, we surveyed about 200 articles related to medical image segmentation, and divided them into three groups based on their attention mechanisms, Pre-Transformer attention, Transformer attention and Mamba-related attention. In each group, we deeply analyze the attention mechanisms from three aspects based on the current literature work, <mml:math altimg=\"si1.svg\" display=\"inline\"><mml:mrow><mml:mi>i</mml:mi><mml:mo>.</mml:mo><mml:mi>e</mml:mi><mml:mo>.</mml:mo></mml:mrow></mml:math>, the principle of the mechanism (what to use), implementation methods (how to use), and application tasks (where to use). We also thoroughly analyzed the advantages and limitations of their applications to different tasks. Finally, we summarize the current state of research and shortcomings in the field, and discuss the potential challenges in the future, including task specificity, robustness, standard evaluation, <ce:italic>etc</ce:italic>. We hope that this review can showcase the overall research context of traditional, Transformer and Mamba attention methods, provide a clear reference for subsequent research, and inspire more advanced attention research, not only in medical image segmentation, but also in other image analysis scenarios. Finally, we maintain the paper list and open-source code at <ce:inter-ref xlink:href=\"https://github.com/Ammexm/Medical-Image-Segmentation\" xlink:type=\"simple\">here</ce:inter-ref>.","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"558 1","pages":""},"PeriodicalIF":12.9,"publicationDate":"2025-01-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142988285","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-11DOI: 10.1016/j.cosrev.2024.100722
S. Ratna, Sukhdeep Singh, Anuj Sharma
Graph Neural Networks (GNNs) have become a prominent technique for the analysis of graph-based data and knowledge extraction. This data can be either structured or unstructured. GNN approaches are particularly beneficial when it comes to examining non-euclidean data. Graph data formats are well-known for their capability to represent intricate systems and understand their relationships. GNNs have significantly advanced the field of research because of their numerous possible applications in machine learning for tasks involving graph-structured data, where relationships between entities play a crucial role. GNN can carry out numerous tasks, such as classifying nodes, categorizing graphs, predicting links or relationships, and much more. Node classification is a widely used and recognized GNN task that has reached state-of-the-art performance on a number of benchmark datasets. In this study, we have provided a comprehensive insight into GNN, its development, and an extensive review of node classification, along with experimental findings and discussions.
{"title":"An inclusive analysis for performance and efficiency of graph neural network models for node classification","authors":"S. Ratna, Sukhdeep Singh, Anuj Sharma","doi":"10.1016/j.cosrev.2024.100722","DOIUrl":"https://doi.org/10.1016/j.cosrev.2024.100722","url":null,"abstract":"Graph Neural Networks (GNNs) have become a prominent technique for the analysis of graph-based data and knowledge extraction. This data can be either structured or unstructured. GNN approaches are particularly beneficial when it comes to examining non-euclidean data. Graph data formats are well-known for their capability to represent intricate systems and understand their relationships. GNNs have significantly advanced the field of research because of their numerous possible applications in machine learning for tasks involving graph-structured data, where relationships between entities play a crucial role. GNN can carry out numerous tasks, such as classifying nodes, categorizing graphs, predicting links or relationships, and much more. Node classification is a widely used and recognized GNN task that has reached state-of-the-art performance on a number of benchmark datasets. In this study, we have provided a comprehensive insight into GNN, its development, and an extensive review of node classification, along with experimental findings and discussions.","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"6 1","pages":""},"PeriodicalIF":12.9,"publicationDate":"2025-01-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142987983","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-09DOI: 10.1016/j.cosrev.2024.100723
Dhanashree Vipul Yevle, Palvinder Singh Mann
Waste management has grown to become one of the leading global challenges due to the massive generation of thousands of tons of waste that is produced daily, leading to severe environmental degradation, the risk of public health, and resource depletion. Despite efforts directed towards solving these problems, traditional methods of sorting and categorizing waste are inefficient and unsustainable, thus requiring the conceptualization of innovative AI-based solutions for more effective waste management. This review presents, a comprehensive review of all the strategies which are critical for AI based techniques, thus improve productivity and sustainability in operations. Diverse datasets used to train AI models along with performance evaluation metrics, and discusses challenges of AI assimilation in waste management systems, most fundamentally the issue of data privacy and concern of bias in the algorithms. Additionally, the role of loss functions and optimizers in enhancing AI model performance and suggests future research opportunities for sustainable resource recovery, recycling, and reuse based on AI.
{"title":"Artificial intelligence based classification for waste management: A survey based on taxonomy, classification & future direction","authors":"Dhanashree Vipul Yevle, Palvinder Singh Mann","doi":"10.1016/j.cosrev.2024.100723","DOIUrl":"https://doi.org/10.1016/j.cosrev.2024.100723","url":null,"abstract":"Waste management has grown to become one of the leading global challenges due to the massive generation of thousands of tons of waste that is produced daily, leading to severe environmental degradation, the risk of public health, and resource depletion. Despite efforts directed towards solving these problems, traditional methods of sorting and categorizing waste are inefficient and unsustainable, thus requiring the conceptualization of innovative AI-based solutions for more effective waste management. This review presents, a comprehensive review of all the strategies which are critical for AI based techniques, thus improve productivity and sustainability in operations. Diverse datasets used to train AI models along with performance evaluation metrics, and discusses challenges of AI assimilation in waste management systems, most fundamentally the issue of data privacy and concern of bias in the algorithms. Additionally, the role of loss functions and optimizers in enhancing AI model performance and suggests future research opportunities for sustainable resource recovery, recycling, and reuse based on AI.","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"6 1","pages":""},"PeriodicalIF":12.9,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142939697","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-09DOI: 10.1016/j.cosrev.2024.100720
Lien P. Le, Thu Nguyen, Michael A. Riegler, Pål Halvorsen, Binh T. Nguyen
The rapid advancement in healthcare data collection technologies and the importance of using multimodal data for accurate diagnosis leads to a surge in multimodal data characterized by different types, structures, and missing values. Machine learning algorithms for predicting or analyzing usually demand the completeness of data. As a result, handling missing data has become a critical concern in the healthcare sector. This survey paper comprehensively reviews recent works on handling multimodal missing data in healthcare. We emphasize methods for synthesizing data from various modalities or multiple sources in imputing missing data, including early fusion, late fusion, and intermediate fusion methods for missing data imputation. The main objective of this study is to identify gaps in the surveyed area and list future tasks and challenges in handling multimodal missing data in healthcare. This review is valuable for researchers and practitioners in healthcare data analysis. It provides insights into using fusion methods to improve data quality and healthcare outcomes.
{"title":"Multimodal missing data in healthcare: A comprehensive review and future directions","authors":"Lien P. Le, Thu Nguyen, Michael A. Riegler, Pål Halvorsen, Binh T. Nguyen","doi":"10.1016/j.cosrev.2024.100720","DOIUrl":"https://doi.org/10.1016/j.cosrev.2024.100720","url":null,"abstract":"The rapid advancement in healthcare data collection technologies and the importance of using multimodal data for accurate diagnosis leads to a surge in multimodal data characterized by different types, structures, and missing values. Machine learning algorithms for predicting or analyzing usually demand the completeness of data. As a result, handling missing data has become a critical concern in the healthcare sector. This survey paper comprehensively reviews recent works on handling multimodal missing data in healthcare. We emphasize methods for synthesizing data from various modalities or multiple sources in imputing missing data, including early fusion, late fusion, and intermediate fusion methods for missing data imputation. The main objective of this study is to identify gaps in the surveyed area and list future tasks and challenges in handling multimodal missing data in healthcare. This review is valuable for researchers and practitioners in healthcare data analysis. It provides insights into using fusion methods to improve data quality and healthcare outcomes.","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"16 1","pages":""},"PeriodicalIF":12.9,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142939737","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-01-06DOI: 10.1016/j.cosrev.2024.100718
Indra Devi K.B., Durai Raj Vincent P.M.
The quest to find reliable biomarkers in autism spectrum disorders (ASD) is an ongoing endeavour to identify both underlying causes and measurable indicators of this neurodevelopmental condition. Machine learning (ML) and advanced deep learning (DL) techniques have enhanced biomarker identification in neuroimaging and behavioral studies, aiding in diagnostic accuracy and early detection. This review paper examines the transformative impact of applying machine learning (ML), particularly deep learning (DL) techniques such as transfer learning and transformer architectures, in advancing ASD diagnosis. The review begins by critically assessing existing literature utilizing ML techniques like logistic regression, random forest, and support vector machines in identifying biomarkers that could potentially aid in the diagnosis of ASD and differentiate between ASD and neurotypical individuals. The focus then shifts to DL models, including Multilayer Perceptrons, Convolutional Neural Networks, Graph Neural Networks, and Long Short-Term Memory networks, to evaluate their suitability for identifying complex patterns linked to ASD. Addressing limited datasets, the review examines transfer learning with pre-trained models, including VGG, ResNet, DenseNet, MobileNet, Inception, and Xception architectures. Additionally, using the ABIDE-I dataset, VGG19, MobileNet, InceptionV3, and DenseNet121 were applied, evaluating their performance through accuracy, sensitivity, specificity, and F1 score. The review further considers transformer architectures, such as Vision Transformers, Swin Transformers, Spatial Temporal Transformers, BolT Transformer, and Convolutional Network Transformer for capturing long-range dependencies in ASD diagnosis. This review aims to be an essential reference for researchers exploring the field of AI-powered ASD diagnosis and classification, by offering analysis of various approaches and highlighting recent advancements.
{"title":"The emergence of artificial intelligence in autism spectrum disorder research: A review of neuro imaging and behavioral applications","authors":"Indra Devi K.B., Durai Raj Vincent P.M.","doi":"10.1016/j.cosrev.2024.100718","DOIUrl":"https://doi.org/10.1016/j.cosrev.2024.100718","url":null,"abstract":"The quest to find reliable biomarkers in autism spectrum disorders (ASD) is an ongoing endeavour to identify both underlying causes and measurable indicators of this neurodevelopmental condition. Machine learning (ML) and advanced deep learning (DL) techniques have enhanced biomarker identification in neuroimaging and behavioral studies, aiding in diagnostic accuracy and early detection. This review paper examines the transformative impact of applying machine learning (ML), particularly deep learning (DL) techniques such as transfer learning and transformer architectures, in advancing ASD diagnosis. The review begins by critically assessing existing literature utilizing ML techniques like logistic regression, random forest, and support vector machines in identifying biomarkers that could potentially aid in the diagnosis of ASD and differentiate between ASD and neurotypical individuals. The focus then shifts to DL models, including Multilayer Perceptrons, Convolutional Neural Networks, Graph Neural Networks, and Long Short-Term Memory networks, to evaluate their suitability for identifying complex patterns linked to ASD. Addressing limited datasets, the review examines transfer learning with pre-trained models, including VGG, ResNet, DenseNet, MobileNet, Inception, and Xception architectures. Additionally, using the ABIDE-I dataset, VGG19, MobileNet, InceptionV3, and DenseNet121 were applied, evaluating their performance through accuracy, sensitivity, specificity, and F1 score. The review further considers transformer architectures, such as Vision Transformers, Swin Transformers, Spatial Temporal Transformers, BolT Transformer, and Convolutional Network Transformer for capturing long-range dependencies in ASD diagnosis. This review aims to be an essential reference for researchers exploring the field of AI-powered ASD diagnosis and classification, by offering analysis of various approaches and highlighting recent advancements.","PeriodicalId":48633,"journal":{"name":"Computer Science Review","volume":"67 1","pages":""},"PeriodicalIF":12.9,"publicationDate":"2025-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142939739","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：