Pub Date : 2024-02-07DOI: 10.1109/ICAIC60265.2024.10433845
Shadi Jawhar, Jeremy Miller, Zeina Bitar
The use of artificial intelligence (AI) in cyber security [1] has proven to be very effective as it helps security professionals better understand, examine, and evaluate possible risks and mitigate them. It also provides guidelines to implement solutions to protect assets and safeguard the technology used. As cyber threats continue to evolve in complexity and scope, and as international standards continuously get updated, the need to generate new policies or update existing ones efficiently and easily has increased [1] [2].The use of (AI) in developing cybersecurity policies and procedures can be key in assuring the correctness and effectiveness of these policies as this is one of the needs for both private organizations and governmental agencies. This study sheds light on the power of AI-driven mechanisms in enhancing digital defense procedures by providing a deep implementation of how AI can aid in generating policies quickly and to the needed level.
{"title":"AI-Based Cybersecurity Policies and Procedures","authors":"Shadi Jawhar, Jeremy Miller, Zeina Bitar","doi":"10.1109/ICAIC60265.2024.10433845","DOIUrl":"https://doi.org/10.1109/ICAIC60265.2024.10433845","url":null,"abstract":"The use of artificial intelligence (AI) in cyber security [1] has proven to be very effective as it helps security professionals better understand, examine, and evaluate possible risks and mitigate them. It also provides guidelines to implement solutions to protect assets and safeguard the technology used. As cyber threats continue to evolve in complexity and scope, and as international standards continuously get updated, the need to generate new policies or update existing ones efficiently and easily has increased [1] [2].The use of (AI) in developing cybersecurity policies and procedures can be key in assuring the correctness and effectiveness of these policies as this is one of the needs for both private organizations and governmental agencies. This study sheds light on the power of AI-driven mechanisms in enhancing digital defense procedures by providing a deep implementation of how AI can aid in generating policies quickly and to the needed level.","PeriodicalId":517265,"journal":{"name":"2024 IEEE 3rd International Conference on AI in Cybersecurity (ICAIC)","volume":"69 1","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2024-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139893342","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-07DOI: 10.1109/ICAIC60265.2024.10433829
Shadi Jawhar, Jeremy Miller, Zeina Bitar
Artificial intelligence (AI) has been successfully used in cyber security for enhancing comprehending, investigating, and evaluating cyber threats. It can effectively anticipate cyber risks in a more efficient way. AI also helps in putting in place strategies to safeguard assets and data. Due to their complexity and constant development, it has been difficult to comprehend cybersecurity controls and adopt the corresponding cyber training and security policies and plans.Given that both cyber academics and cyber practitioners need to have a deep comprehension of cybersecurity rules, artificial intelligence (AI) in cybersecurity can be a crucial tool in both education and awareness. By offering an in-depth demonstration of how AI may help in cybersecurity education and awareness and in creating policies fast and to the needed level, this study focuses on the efficiency of AI-driven mechanisms in strengthening the entire cyber security education life cycle.
{"title":"AI-Driven Customized Cyber Security Training and Awareness","authors":"Shadi Jawhar, Jeremy Miller, Zeina Bitar","doi":"10.1109/ICAIC60265.2024.10433829","DOIUrl":"https://doi.org/10.1109/ICAIC60265.2024.10433829","url":null,"abstract":"Artificial intelligence (AI) has been successfully used in cyber security for enhancing comprehending, investigating, and evaluating cyber threats. It can effectively anticipate cyber risks in a more efficient way. AI also helps in putting in place strategies to safeguard assets and data. Due to their complexity and constant development, it has been difficult to comprehend cybersecurity controls and adopt the corresponding cyber training and security policies and plans.Given that both cyber academics and cyber practitioners need to have a deep comprehension of cybersecurity rules, artificial intelligence (AI) in cybersecurity can be a crucial tool in both education and awareness. By offering an in-depth demonstration of how AI may help in cybersecurity education and awareness and in creating policies fast and to the needed level, this study focuses on the efficiency of AI-driven mechanisms in strengthening the entire cyber security education life cycle.","PeriodicalId":517265,"journal":{"name":"2024 IEEE 3rd International Conference on AI in Cybersecurity (ICAIC)","volume":"73 12","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2024-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139895393","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-07DOI: 10.1109/ICAIC60265.2024.10433845
Shadi Jawhar, Jeremy Miller, Zeina Bitar
The use of artificial intelligence (AI) in cyber security [1] has proven to be very effective as it helps security professionals better understand, examine, and evaluate possible risks and mitigate them. It also provides guidelines to implement solutions to protect assets and safeguard the technology used. As cyber threats continue to evolve in complexity and scope, and as international standards continuously get updated, the need to generate new policies or update existing ones efficiently and easily has increased [1] [2].The use of (AI) in developing cybersecurity policies and procedures can be key in assuring the correctness and effectiveness of these policies as this is one of the needs for both private organizations and governmental agencies. This study sheds light on the power of AI-driven mechanisms in enhancing digital defense procedures by providing a deep implementation of how AI can aid in generating policies quickly and to the needed level.
{"title":"AI-Based Cybersecurity Policies and Procedures","authors":"Shadi Jawhar, Jeremy Miller, Zeina Bitar","doi":"10.1109/ICAIC60265.2024.10433845","DOIUrl":"https://doi.org/10.1109/ICAIC60265.2024.10433845","url":null,"abstract":"The use of artificial intelligence (AI) in cyber security [1] has proven to be very effective as it helps security professionals better understand, examine, and evaluate possible risks and mitigate them. It also provides guidelines to implement solutions to protect assets and safeguard the technology used. As cyber threats continue to evolve in complexity and scope, and as international standards continuously get updated, the need to generate new policies or update existing ones efficiently and easily has increased [1] [2].The use of (AI) in developing cybersecurity policies and procedures can be key in assuring the correctness and effectiveness of these policies as this is one of the needs for both private organizations and governmental agencies. This study sheds light on the power of AI-driven mechanisms in enhancing digital defense procedures by providing a deep implementation of how AI can aid in generating policies quickly and to the needed level.","PeriodicalId":517265,"journal":{"name":"2024 IEEE 3rd International Conference on AI in Cybersecurity (ICAIC)","volume":"16 5","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2024-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139895506","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-07DOI: 10.1109/ICAIC60265.2024.10433838
C. Mutongi, Billy Rigava
Today one has to run very fast to stay on the same position. We are no longer competing with humans only, we are now also competing with robots as they are involved in learning, leading to Machine Learning (ML). Robots are increasingly being adopted in healthcare to carry out various tasks that enhance patient care. Robots in health care have revolutionized the health ecosystem. There are different types of healthcare robots which include nursing robots, surgical robots, clinical Training, Prescription Dispensing, care robots, Telepresence, Rehabilitation Robots, Health Call Centre Robots, Ambulance Robots and Physical Therapy Robots. Healthcare robots are mostly found in the developed countries. This paper seeks to establish robotics in healthcare considering the African perspectives and Zimbabwe in particular. A qualitative study was conducted whereby twenty students at a university were interviewed concerning their views on healthcare robots in the African context. It was found out that healthcare robots are still at their conception in Africa and Zimbabwe in particular, there is fear of the unknown, some indicated that robots will affect their indigenous way of life as they are used to interact with each other as human beings and not as robot to human as shown by the concept of Ubuntu, power challenges, connectivity, lack of awareness challenges, as well as cultural and religious challenges. However, some participants indicated that they greatly welcome the robots as they may cease the health professional shortages in Africa and also they consider them to be more precise and accurate as compared to humans. Some indicated that more privacy will be promoted due to the use of robots. It was recommended that there is need for immense healthcare robots conscientisation, awareness, training, robots to mimic the African way of living and language.
{"title":"Robotics in Healthcare: The African Perspective","authors":"C. Mutongi, Billy Rigava","doi":"10.1109/ICAIC60265.2024.10433838","DOIUrl":"https://doi.org/10.1109/ICAIC60265.2024.10433838","url":null,"abstract":"Today one has to run very fast to stay on the same position. We are no longer competing with humans only, we are now also competing with robots as they are involved in learning, leading to Machine Learning (ML). Robots are increasingly being adopted in healthcare to carry out various tasks that enhance patient care. Robots in health care have revolutionized the health ecosystem. There are different types of healthcare robots which include nursing robots, surgical robots, clinical Training, Prescription Dispensing, care robots, Telepresence, Rehabilitation Robots, Health Call Centre Robots, Ambulance Robots and Physical Therapy Robots. Healthcare robots are mostly found in the developed countries. This paper seeks to establish robotics in healthcare considering the African perspectives and Zimbabwe in particular. A qualitative study was conducted whereby twenty students at a university were interviewed concerning their views on healthcare robots in the African context. It was found out that healthcare robots are still at their conception in Africa and Zimbabwe in particular, there is fear of the unknown, some indicated that robots will affect their indigenous way of life as they are used to interact with each other as human beings and not as robot to human as shown by the concept of Ubuntu, power challenges, connectivity, lack of awareness challenges, as well as cultural and religious challenges. However, some participants indicated that they greatly welcome the robots as they may cease the health professional shortages in Africa and also they consider them to be more precise and accurate as compared to humans. Some indicated that more privacy will be promoted due to the use of robots. It was recommended that there is need for immense healthcare robots conscientisation, awareness, training, robots to mimic the African way of living and language.","PeriodicalId":517265,"journal":{"name":"2024 IEEE 3rd International Conference on AI in Cybersecurity (ICAIC)","volume":"1 1","pages":"1-10"},"PeriodicalIF":0.0,"publicationDate":"2024-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139895510","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-07DOI: 10.1109/ICAIC60265.2024.10433843
Gideon Popoola, Khadijat-Kuburat Abdullah, Gerard Shu Fuhnwi, Janet O. Agbaje
Blogs, online forums, comment sections, and social networking sites like Facebook, Twitter (now known as X), and Instagram can all be called social media. The growing use of social media has made some unstructured data available, which can benefit us if we clean, structure, and analyze the data. Twitter is a popular microblogging social media platform where people share and express their opinions about any topic. The act of analyzing these opinions of people is called sentimental analysis. Sentimental analysis can be helpful to individuals, businesses, government agencies, etc. In this study, tweets related to financial news were extracted, labeled, and analyzed to capture the opinions of people around the world. This paper proposes a novel machine learning-based approach to analyze social media data for sentiment analysis. The presented approach is divided into three steps. The first stage is preprocessing, where the tweets are refined and filtered. In the second stage, feature extraction was performed using Term Frequency and Inverse Document Frequency (TF-IDF). The third stage involves using the extracted features to make predictions using machine learning algorithms. Three machine learning models were used, namely, random forest classifier (RF), Naïve Bayes (NB), and k-nearest neighbor (KNN). The evaluation results show that both NB and RF perform better than KNN in accuracy, precision, Recall, and F1-score metrics. These results also show an overwhelmingly positive opinion regarding financial news.
博客、在线论坛、评论区以及 Facebook、Twitter（现在称为 X）和 Instagram 等社交网站都可称为社交媒体。社交媒体的使用日益增多，使得一些非结构化数据变得可用，如果我们对这些数据进行清理、结构化和分析，就能从中受益。Twitter 是一个流行的微博社交媒体平台，人们在这个平台上分享和表达自己对任何话题的看法。对这些观点进行分析的行为被称为情感分析。情感分析对个人、企业、政府机构等都有帮助。本研究对与财经新闻相关的推文进行了提取、标记和分析，以捕捉世界各地人们的观点。本文提出了一种基于机器学习的新方法来分析社交媒体数据，以进行情感分析。该方法分为三个步骤。第一阶段是预处理，对推文进行提炼和过滤。在第二阶段，使用术语频率和反向文档频率（TF-IDF）进行特征提取。第三阶段是利用提取的特征，使用机器学习算法进行预测。使用了三种机器学习模型，即随机森林分类器（RF）、奈夫贝叶斯（NB）和 k 近邻（KNN）。评估结果表明，NB 和 RF 在准确率、精确度、召回率和 F1 分数指标上都优于 KNN。这些结果还表明，人们对财经新闻的看法绝大多数是正面的。
{"title":"Sentiment Analysis of Financial News Data using TF-IDF and Machine Learning Algorithms","authors":"Gideon Popoola, Khadijat-Kuburat Abdullah, Gerard Shu Fuhnwi, Janet O. Agbaje","doi":"10.1109/ICAIC60265.2024.10433843","DOIUrl":"https://doi.org/10.1109/ICAIC60265.2024.10433843","url":null,"abstract":"Blogs, online forums, comment sections, and social networking sites like Facebook, Twitter (now known as X), and Instagram can all be called social media. The growing use of social media has made some unstructured data available, which can benefit us if we clean, structure, and analyze the data. Twitter is a popular microblogging social media platform where people share and express their opinions about any topic. The act of analyzing these opinions of people is called sentimental analysis. Sentimental analysis can be helpful to individuals, businesses, government agencies, etc. In this study, tweets related to financial news were extracted, labeled, and analyzed to capture the opinions of people around the world. This paper proposes a novel machine learning-based approach to analyze social media data for sentiment analysis. The presented approach is divided into three steps. The first stage is preprocessing, where the tweets are refined and filtered. In the second stage, feature extraction was performed using Term Frequency and Inverse Document Frequency (TF-IDF). The third stage involves using the extracted features to make predictions using machine learning algorithms. Three machine learning models were used, namely, random forest classifier (RF), Naïve Bayes (NB), and k-nearest neighbor (KNN). The evaluation results show that both NB and RF perform better than KNN in accuracy, precision, Recall, and F1-score metrics. These results also show an overwhelmingly positive opinion regarding financial news.","PeriodicalId":517265,"journal":{"name":"2024 IEEE 3rd International Conference on AI in Cybersecurity (ICAIC)","volume":"283 8","pages":"1-6"},"PeriodicalIF":0.0,"publicationDate":"2024-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139896070","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-07DOI: 10.1109/ICAIC60265.2024.10433800
M. Raval, Jin Aobo, Yun Wan, Hardik A. Gohel
Pneumonia prediction using chest X-ray images is a challenging task because of the complex image processing involved. The radiographic features of pneumonia, especially in the earlier stages, easily overlap with other lung conditions, which makes the differentiation even more challenging. Moreover, X-ray image quality varies due to equipment, patient condition, and techniques, particularly in rural areas with undertrained radiologists and medical experts. The use of Artificial Intelligence (AI) models in detecting pneumonia is a novel but crucial research field and rapid advancement in medical imaging technology and neural network models along with the availability of large de-identified public datasets has paved the way for this life-saving biomedical research. In this paper, we propose a unique comprehensive solution for predicting pneumonia using chest X-ray images. We utilize an enhanced VGGNet model tailored for the binary classification task. The modified VGG19 with a binary classifier provides a solid foundation for feature extraction and leverages the pretrained features and deep architecture to differentiate between normal and pneumonia-affected lung images. The use of transfer learning allows us to extend the pre-trained model on a diverse and large-scale dataset by further training it on limited-size medical imaging data for the crucial task of biomedical classification without the need for large, labeled training data or computational resources. The robust model displays high accuracy of 92% with a high recall of 96.4% and AUC of 97%. With high adaptability and efficient learning capacity from limited data. This implementation may serve as a powerful tool assisting medical professionals in diagnosing pneumonia by quickly analyzing X-ray images with the same consistency and accuracy. During crises such as pandemics where lung diseases might surge, such tools can aid in rapid screening and monitoring of public health.
由于涉及复杂的图像处理，使用胸部 X 光图像预测肺炎是一项具有挑战性的任务。肺炎的影像学特征，尤其是早期肺炎的影像学特征，很容易与其他肺部疾病重叠，这使得区分肺炎的工作更具挑战性。此外，X 射线图像质量因设备、患者状况和技术而异，尤其是在农村地区，放射科医生和医疗专家的培训不足。人工智能（AI）模型在肺炎检测中的应用是一个新颖而关键的研究领域，医学成像技术和神经网络模型的快速发展以及大量去标识化公共数据集的可用性为这一拯救生命的生物医学研究铺平了道路。在本文中，我们提出了利用胸部 X 光图像预测肺炎的独特综合解决方案。我们采用了专为二元分类任务定制的增强型 VGGNet 模型。带有二元分类器的改进型 VGG19 为特征提取奠定了坚实的基础，并利用预训练特征和深度架构来区分正常肺部图像和受肺炎影响的肺部图像。迁移学习的使用使我们能够通过在有限规模的医学影像数据上进一步训练预训练模型，从而在多样化的大规模数据集上扩展预训练模型，以完成生物医学分类的关键任务，而无需大量标注训练数据或计算资源。该稳健模型的准确率高达 92%，召回率高达 96.4%，AUC 高达 97%。该模型适应性强，能从有限的数据中高效学习。通过快速分析具有相同一致性和准确性的 X 光图像，该实施方案可作为协助医疗专业人员诊断肺炎的有力工具。在肺部疾病可能激增的大流行等危机期间，这种工具可以帮助快速筛查和监测公共卫生。
{"title":"Leveraging Advanced Visual Recognition Classifier For Pneumonia Prediction","authors":"M. Raval, Jin Aobo, Yun Wan, Hardik A. Gohel","doi":"10.1109/ICAIC60265.2024.10433800","DOIUrl":"https://doi.org/10.1109/ICAIC60265.2024.10433800","url":null,"abstract":"Pneumonia prediction using chest X-ray images is a challenging task because of the complex image processing involved. The radiographic features of pneumonia, especially in the earlier stages, easily overlap with other lung conditions, which makes the differentiation even more challenging. Moreover, X-ray image quality varies due to equipment, patient condition, and techniques, particularly in rural areas with undertrained radiologists and medical experts. The use of Artificial Intelligence (AI) models in detecting pneumonia is a novel but crucial research field and rapid advancement in medical imaging technology and neural network models along with the availability of large de-identified public datasets has paved the way for this life-saving biomedical research. In this paper, we propose a unique comprehensive solution for predicting pneumonia using chest X-ray images. We utilize an enhanced VGGNet model tailored for the binary classification task. The modified VGG19 with a binary classifier provides a solid foundation for feature extraction and leverages the pretrained features and deep architecture to differentiate between normal and pneumonia-affected lung images. The use of transfer learning allows us to extend the pre-trained model on a diverse and large-scale dataset by further training it on limited-size medical imaging data for the crucial task of biomedical classification without the need for large, labeled training data or computational resources. The robust model displays high accuracy of 92% with a high recall of 96.4% and AUC of 97%. With high adaptability and efficient learning capacity from limited data. This implementation may serve as a powerful tool assisting medical professionals in diagnosing pneumonia by quickly analyzing X-ray images with the same consistency and accuracy. During crises such as pandemics where lung diseases might surge, such tools can aid in rapid screening and monitoring of public health.","PeriodicalId":517265,"journal":{"name":"2024 IEEE 3rd International Conference on AI in Cybersecurity (ICAIC)","volume":"30 11‐12","pages":"1-8"},"PeriodicalIF":0.0,"publicationDate":"2024-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139893349","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The main objective of this project is to increase the productivity of farmers producing fruits and vegetables in Kazakhstan. We are planning to use technology during production at orchards and also using technology during storage. At the production stage, we shall capture images of fruit flowers, growing fruits, and a ripe fruit. Then we shall apply federated learning to train our model with healthy fruits and flowers and then we shall be able to predict any ongoing pest infections with either fruits or flower. At the storage phase, we shall use e-nose to check the current status of apple and save it from any possible degradation. We shall also use blockchain to store data related to fruits at both stages of production and storage to create an e-passport that will give access to data related to production and storage of fruits. At the same time, we shall also use various width clustering algorithms to detect intrusion in our sensor based IoT networks.
{"title":"Federated Learning Based Smart Horticulture and Smart Storage of Fruits Using E-Nose, and Blockchain: A Proposed Model","authors":"Shakhmaran Seilov, Bishwajeet Pandey, Akniyet Nurzhaubayev, Dias Abildinov, Assem Konyrkhanova, Bibinur Zhursinbek","doi":"10.1109/ICAIC60265.2024.10433834","DOIUrl":"https://doi.org/10.1109/ICAIC60265.2024.10433834","url":null,"abstract":"The main objective of this project is to increase the productivity of farmers producing fruits and vegetables in Kazakhstan. We are planning to use technology during production at orchards and also using technology during storage. At the production stage, we shall capture images of fruit flowers, growing fruits, and a ripe fruit. Then we shall apply federated learning to train our model with healthy fruits and flowers and then we shall be able to predict any ongoing pest infections with either fruits or flower. At the storage phase, we shall use e-nose to check the current status of apple and save it from any possible degradation. We shall also use blockchain to store data related to fruits at both stages of production and storage to create an e-passport that will give access to data related to production and storage of fruits. At the same time, we shall also use various width clustering algorithms to detect intrusion in our sensor based IoT networks.","PeriodicalId":517265,"journal":{"name":"2024 IEEE 3rd International Conference on AI in Cybersecurity (ICAIC)","volume":"78 3-4","pages":"1-5"},"PeriodicalIF":0.0,"publicationDate":"2024-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139895903","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-07DOI: 10.1109/ICAIC60265.2024.10433800
M. Raval, Jin Aobo, Yun Wan, Hardik A. Gohel
Pneumonia prediction using chest X-ray images is a challenging task because of the complex image processing involved. The radiographic features of pneumonia, especially in the earlier stages, easily overlap with other lung conditions, which makes the differentiation even more challenging. Moreover, X-ray image quality varies due to equipment, patient condition, and techniques, particularly in rural areas with undertrained radiologists and medical experts. The use of Artificial Intelligence (AI) models in detecting pneumonia is a novel but crucial research field and rapid advancement in medical imaging technology and neural network models along with the availability of large de-identified public datasets has paved the way for this life-saving biomedical research. In this paper, we propose a unique comprehensive solution for predicting pneumonia using chest X-ray images. We utilize an enhanced VGGNet model tailored for the binary classification task. The modified VGG19 with a binary classifier provides a solid foundation for feature extraction and leverages the pretrained features and deep architecture to differentiate between normal and pneumonia-affected lung images. The use of transfer learning allows us to extend the pre-trained model on a diverse and large-scale dataset by further training it on limited-size medical imaging data for the crucial task of biomedical classification without the need for large, labeled training data or computational resources. The robust model displays high accuracy of 92% with a high recall of 96.4% and AUC of 97%. With high adaptability and efficient learning capacity from limited data. This implementation may serve as a powerful tool assisting medical professionals in diagnosing pneumonia by quickly analyzing X-ray images with the same consistency and accuracy. During crises such as pandemics where lung diseases might surge, such tools can aid in rapid screening and monitoring of public health.
由于涉及复杂的图像处理，使用胸部 X 光图像预测肺炎是一项具有挑战性的任务。肺炎的影像学特征，尤其是早期肺炎的影像学特征，很容易与其他肺部疾病重叠，这使得区分肺炎的工作更具挑战性。此外，X 射线图像质量因设备、患者状况和技术而异，尤其是在农村地区，放射科医生和医疗专家的培训不足。人工智能（AI）模型在肺炎检测中的应用是一个新颖而关键的研究领域，医学成像技术和神经网络模型的快速发展以及大量去标识化公共数据集的可用性为这一拯救生命的生物医学研究铺平了道路。在本文中，我们提出了利用胸部 X 光图像预测肺炎的独特综合解决方案。我们采用了专为二元分类任务定制的增强型 VGGNet 模型。带有二元分类器的改进型 VGG19 为特征提取奠定了坚实的基础，并利用预训练特征和深度架构来区分正常肺部图像和受肺炎影响的肺部图像。迁移学习的使用使我们能够通过在有限规模的医学影像数据上进一步训练预训练模型，从而在多样化的大规模数据集上扩展预训练模型，以完成生物医学分类的关键任务，而无需大量标注训练数据或计算资源。该稳健模型的准确率高达 92%，召回率高达 96.4%，AUC 高达 97%。该模型适应性强，能从有限的数据中高效学习。通过快速分析具有相同一致性和准确性的 X 光图像，该实施方案可作为协助医疗专业人员诊断肺炎的有力工具。在肺部疾病可能激增的大流行等危机期间，这种工具可以帮助快速筛查和监测公共卫生。
{"title":"Leveraging Advanced Visual Recognition Classifier For Pneumonia Prediction","authors":"M. Raval, Jin Aobo, Yun Wan, Hardik A. Gohel","doi":"10.1109/ICAIC60265.2024.10433800","DOIUrl":"https://doi.org/10.1109/ICAIC60265.2024.10433800","url":null,"abstract":"Pneumonia prediction using chest X-ray images is a challenging task because of the complex image processing involved. The radiographic features of pneumonia, especially in the earlier stages, easily overlap with other lung conditions, which makes the differentiation even more challenging. Moreover, X-ray image quality varies due to equipment, patient condition, and techniques, particularly in rural areas with undertrained radiologists and medical experts. The use of Artificial Intelligence (AI) models in detecting pneumonia is a novel but crucial research field and rapid advancement in medical imaging technology and neural network models along with the availability of large de-identified public datasets has paved the way for this life-saving biomedical research. In this paper, we propose a unique comprehensive solution for predicting pneumonia using chest X-ray images. We utilize an enhanced VGGNet model tailored for the binary classification task. The modified VGG19 with a binary classifier provides a solid foundation for feature extraction and leverages the pretrained features and deep architecture to differentiate between normal and pneumonia-affected lung images. The use of transfer learning allows us to extend the pre-trained model on a diverse and large-scale dataset by further training it on limited-size medical imaging data for the crucial task of biomedical classification without the need for large, labeled training data or computational resources. The robust model displays high accuracy of 92% with a high recall of 96.4% and AUC of 97%. With high adaptability and efficient learning capacity from limited data. This implementation may serve as a powerful tool assisting medical professionals in diagnosing pneumonia by quickly analyzing X-ray images with the same consistency and accuracy. During crises such as pandemics where lung diseases might surge, such tools can aid in rapid screening and monitoring of public health.","PeriodicalId":517265,"journal":{"name":"2024 IEEE 3rd International Conference on AI in Cybersecurity (ICAIC)","volume":"10 11","pages":"1-8"},"PeriodicalIF":0.0,"publicationDate":"2024-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139896051","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-07DOI: 10.1109/ICAIC60265.2024.10433836
Xavier Lessage, Leandro Collier, Charles-Henry Bertrand Van Ouytsel, Axel Legay, Saïd Mahmoudi, Philippe Massonet
This study explores the convergence of Federated Learning (FL) and Fully Homomorphic Encryption (FHE) through an innovative approach applied to a confidential dataset composed of mammograms from Belgian medical records. Our goal is to clarify the feasibility and challenges associated with integrating FHE into the context of Federated Learning, with a particular focus on evaluating the memory constraints inherent in FHE when using sensitive medical data. The results highlight notable limitations in terms of memory usage, underscoring the need for ongoing research to optimize FHE in real-world applications. Despite these challenges, our research demonstrates that FHE maintains comparable performance in terms of Receiver Operating Characteristic (ROC) curves, affirming the robustness of our approach in secure machine learning applications, especially in sectors where data confidentiality, such as medical data management, is imperative. The conclusions not only shed light on the technical limitations of FHE but also emphasize its potential for practical applications. By combining Federated Learning with FHE, our model preserves data confidentiality while ensuring the security of exchanges between participants and the central server
{"title":"Secure federated learning applied to medical imaging with fully homomorphic encryption","authors":"Xavier Lessage, Leandro Collier, Charles-Henry Bertrand Van Ouytsel, Axel Legay, Saïd Mahmoudi, Philippe Massonet","doi":"10.1109/ICAIC60265.2024.10433836","DOIUrl":"https://doi.org/10.1109/ICAIC60265.2024.10433836","url":null,"abstract":"This study explores the convergence of Federated Learning (FL) and Fully Homomorphic Encryption (FHE) through an innovative approach applied to a confidential dataset composed of mammograms from Belgian medical records. Our goal is to clarify the feasibility and challenges associated with integrating FHE into the context of Federated Learning, with a particular focus on evaluating the memory constraints inherent in FHE when using sensitive medical data. The results highlight notable limitations in terms of memory usage, underscoring the need for ongoing research to optimize FHE in real-world applications. Despite these challenges, our research demonstrates that FHE maintains comparable performance in terms of Receiver Operating Characteristic (ROC) curves, affirming the robustness of our approach in secure machine learning applications, especially in sectors where data confidentiality, such as medical data management, is imperative. The conclusions not only shed light on the technical limitations of FHE but also emphasize its potential for practical applications. By combining Federated Learning with FHE, our model preserves data confidentiality while ensuring the security of exchanges between participants and the central server","PeriodicalId":517265,"journal":{"name":"2024 IEEE 3rd International Conference on AI in Cybersecurity (ICAIC)","volume":"259 1","pages":"1-12"},"PeriodicalIF":0.0,"publicationDate":"2024-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139896084","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2024-02-07DOI: 10.1109/ICAIC60265.2024.10433827
Antonio Maci, Nicola Tamma, Anthony J. Coscia
Data theft through web applications that emulate legitimate platforms constitutes a major network security issue. Countermeasures using artificial intelligence (AI)-based systems are often applied because they can effectively detect malicious websites, which are extremely outnumbered by legitimate ones. In this domain, deep reinforcement learning (DRL) emerges as an attractive field for the development of network intrusion detection models, even in the case of highly skewed class distributions. However, DRL requires training time that increases with data complexity. This paper combines a DRL-based classifier with state-of-the-art feature selection techniques to speed up training while retaining or even improving classification performance. Our experiments used the Mendeley dataset and five different statistical and correlation-based feature-ranking strategies. The results indicated that the selection technique based on the calculation of the Gini index reduces the number of columns in the dataset by 27%, saving more than 10% of training time and significantly improving classification scores compared with the case without selection strategies.
{"title":"Deep Reinforcement Learning-based Malicious URL Detection with Feature Selection","authors":"Antonio Maci, Nicola Tamma, Anthony J. Coscia","doi":"10.1109/ICAIC60265.2024.10433827","DOIUrl":"https://doi.org/10.1109/ICAIC60265.2024.10433827","url":null,"abstract":"Data theft through web applications that emulate legitimate platforms constitutes a major network security issue. Countermeasures using artificial intelligence (AI)-based systems are often applied because they can effectively detect malicious websites, which are extremely outnumbered by legitimate ones. In this domain, deep reinforcement learning (DRL) emerges as an attractive field for the development of network intrusion detection models, even in the case of highly skewed class distributions. However, DRL requires training time that increases with data complexity. This paper combines a DRL-based classifier with state-of-the-art feature selection techniques to speed up training while retaining or even improving classification performance. Our experiments used the Mendeley dataset and five different statistical and correlation-based feature-ranking strategies. The results indicated that the selection technique based on the calculation of the Gini index reduces the number of columns in the dataset by 27%, saving more than 10% of training time and significantly improving classification scores compared with the case without selection strategies.","PeriodicalId":517265,"journal":{"name":"2024 IEEE 3rd International Conference on AI in Cybersecurity (ICAIC)","volume":"6 3","pages":"1-7"},"PeriodicalIF":0.0,"publicationDate":"2024-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139895703","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","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：