Yazar "Jamil, A." seçeneğine göre listele
Listeleniyor 1 - 10 / 10
Sayfa Başına Sonuç
Sıralama seçenekleri
Öğe Automated and Optimised Machine Learning Algorithms for Healthcare Informatics(Springer Science and Business Media Deutschland GmbH, 2024) Juyal, A.; Bhushan, B.; Hameed, A.A.; Jamil, A.; Pandey, S.Healthcare is a rapidly expanding field with a substantial amount of heterogeneous data driving innumerable health-related tasks. Various healthcare service providers still rely on manual procedures, which can be time-consuming and require significant effort. To automate such manual operations, recent technological advances have emerged in the domain of Machine Learning (ML). ML falls under the subject of Artificial Intelligence (AI), and it gets combined with ‘big data’ to draw meaningful insights. The integration of ML in the healthcare sector has optimized decision-making and predictive analysis. This paper discusses the various application areas of ML in healthcare. Additionally, several ML algorithms used by other researchers in healthcare-related experiments are summarized. A brief review is provided regarding the experiments. This paper delineates the challenges associated with using ML in healthcare. Finally, the paper offers insights into future research directions. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.Öğe Blockchain Based Security Framework for Internet of Medical Things (IoMT) Applications(Association for Computing Machinery, 2023) Baniya, P.; Nand, P.; Bhushan, B.; Hameed, A.A.; Jamil, A.The growing demand for Internet of Medical Things (IoMT) sensors within the healthcare sector has emphasized the necessity for robust security measures. However, these devices, often constrained by limited resources, frequently require external assistance to guarantee the security of data. Blockchain technology emerges as a practical solution, offering a way to enhance system security and safeguard patients' sensitive health records against unauthorized access and data breaches. The fusion of IoMT and blockchain presents a range of benefits, such as cost reduction, improved speed, automation, and the guarantee of unmodifiable data. Nevertheless, security concerns persist, particularly in light of the ever-evolving encryption techniques. The effective implementation of IoMT hinges on the delivery of personalized, patient-centered care, which has the potential to enhance affordability, simplicity, and overall convenience, ultimately contributing to improved quality of life and extended lifespans. In this comprehensive exploration, we delve into the historical context, architecture, components, and challenges associated with IoMT. Additionally, we conduct an in-depth examination of blockchain technology, covering its essential elements and categories, as well as its seamless integration into IoMT, underscoring its critical role in reinforcing security measures within the healthcare domain. Furthermore, this paper outlines recent developments in blockchain technology within the IoMT framework, highlighting its ongoing evolution and sustained significance in the healthcare field. © 2023 ACM.Öğe Clinical Decision Support System for Diabetes Classification with an Optimized CNN using PSO(Institute of Electrical and Electronics Engineers Inc., 2023) Khan, F.A.; Jamil, A.; Hameed, A.A.; Moetesum, M.Diabetes, a pervasive chronic metabolic disorder, affects a substantial portion of the global population. The timely and precise diagnosis of diabetes is pivotal for effective management and improved patient outcomes. In this study, we introduce an innovative approach aimed at augmenting the classification performance of Convolutional Neural Networks (CNNs) in diabetes diagnosis. Utilizing the Particle Swarm Optimization (PSO) algorithm, we fine-tune the CNN model to enhance the accuracy and efficiency of diabetes classification. Our comprehensive experiments, conducted on the Pima Indians Diabetes dataset, substantiate the effectiveness of our optimized CNN. These findings underscore the potential of integrating CNN and PSO optimization methodologies to significantly boost the accuracy of diabetes diagnosis, thereby facilitating more accurate assessments and tailored treatment strategies for patients. We evaluated the proposed model using standard metrics such as precision, recall, F1-score, and overall accuracy, with results demonstrating that the PSO-based optimized CNN model outperforms the custom CNN, achieving the highest precision, recall, F1-score, and overall accuracy. © 2023 IEEE.Öğe Deep Learning Approaches for Cyber Threat Detection and Mitigation(Association for Computing Machinery, 2023) Juyal, A.; Bhushan, B.; Hameed, A.A.; Jamil, A.Cyberspace has inflated over the past decade, primarily driven by pervasive development and widespread usage of the internet. Prolonged cyber-attacks and security vulnerabilities have become more common as a consequence. A recent study conducted by Ponemon Institute revealed that 64% of businesses have suffered web-based attacks in 2023. Thus, cybersecurity measures have become mandatory in every sector, that leverages technology, to protect confidential information from malignant threats. The majority of the sectors use conventional security measures like anti-virus products to defend their systems against outside attacks, however, these products fail to work on unseen and polymorphic threats. With cutting-edge deep learning (DL) processes getting adopted in nearly every field, it has been widely exploited in cybersecurity to detect and classify cyber-attacks. This study discusses the application of common deep learning models in cybersecurity to detect malware, intrusion, phishes, and spam. This study provides a description of deep learning models and their mathematical backgrounds that are common in cybersecurity. This study also reviews and analyses the work of other researchers who experimented with deep learning models in cybersecurity. This study delineates the challenges faced by deep learning models in cybersecurity. Finally, concluding remarks and future research directions are summarized to foresee the potential improvement of deep learning in cybersecurity. © 2023 ACM.Öğe Enhancing the Multiclass Image Classification Accuracy using Binary Classifiers for Semi-Supervised Learning(Institute of Electrical and Electronics Engineers Inc., 2023) Jadoon, H.K.; Jamil, A.; Zulfiqar, A.; Hameed, A.A.Image classification poses a fundamental challenge in deep learning, especially in scenarios where labeled data is scarce but unlabeled data is abundant. Precise pseudo-labels are crucial to facilitate classification in such situations. One common approach involves the use of binary classifiers with a one-vs-all strategy to assign pseudo-labels to unlabeled data, offering the advantage of tailored predictions for each class. However, this method faces challenges, including class imbalance, often requiring oversampling for resolution, and extended training times due to multiple binary classifiers. Our proposed approach addresses the inherent class imbalance in the one-vs-all method, eliminating the need for oversampling. We achieve this by training a single multi-class classifier through a combination of binary classifiers, transfer learning, and fine-tuning while enforcing a stringent prediction threshold for pseudo-labels. This transition to a single multi-class classifier significantly reduces both training duration and storage demands. Our model's effectiveness is rigorously evaluated on two diverse datasets, MNIST, and Fashion MNIST, achieving impressive test accuracies of 95.59% and 84.84%, respectively, for a pseudo-label generation. © 2023 IEEE.Öğe Implementation of Homomorphic Encryption Schemes in Fog Computing(Springer Science and Business Media Deutschland GmbH, 2024) Pandey, S.; Bhushan, B.; Hameed, A.A.; Jamil, A.; Juyal, A.In today's date, fog computing is on the rise due to its characteristics. Its ability to provide smart devices with close-by computing capabilities offers a drastic reduction in cloud traffic and more efficient data transfer. However, data security in fog computing devices is a major concern. Homomorphic Encryption (HME) Scheme is a technique that protects private data from various threats, and to improve efficiency and reduce time, it offers modification of original data after encryption without decrypting it. In this paper, we explore the various homomorphic encryption schemes that can be implemented in fog computing to protect the data and provide data security. We present a detailed architecture and various characteristics of fog computing to better visualize the use of homomorphic encryption in different areas. Further, we present the applications of homomorphic encryption in different sectors and discuss future research directions. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.Öğe A Multi-modal Approach to Lung Tumor Detection using Deep Learning(Institute of Electrical and Electronics Engineers Inc., 2023) Zafar, A.; Muneeb, S.; Amir, M.; Jamil, A.; Hameed, A.A.Lung cancer remains a significant global cause of cancer-related deaths, emphasizing the importance of early detection for improving patient survival rates. This paper introduces an enhanced approach that aims to achieve efficient and precise lung tumor detection and segmentation. The proposed method utilizes a multimodal approach by leveraging both CT and PET scans, enabling improved tumor detection. The methodology incorporates state-of-The-Art deep learning architectures, including ResNet, DenseNet, and Inception-v3, for effective tumor classification. Additionally, both immediate fusion (early fusion) and late fusion techniques are applied to integrate data from multiple modalities. The performance of the classification models is evaluated using metrics such as precision, F1 score, accuracy, and sensitivity. The experimental results demonstrate the effectiveness of the proposed approach in accurately segmenting lung tumors. The findings contribute to the existing knowledge in the field of tumor segmentation and medical image analysis, providing valuable insights into the benefits of multimodal fusion and deep learning techniques for lung cancer diagnosis and treatment planning. © 2023 IEEE.Öğe Photovoltaics Cell Anomaly Detection Using Deep Learning Techniques(Springer Science and Business Media Deutschland GmbH, 2024) Al-Dulaimi, A.A.; Hameed, A.A.; Guneser, M.T.; Jamil, A.Photovoltaic cells play a crucial role in converting sunlight into electrical energy. However, defects can occur during the manufacturing process, negatively impacting these cells’ efficiency and overall performance. Electroluminescence (EL) imaging has emerged as a viable method for defect detection in photovoltaic cells. Developing an accurate and automated detection model capable of identifying and classifying defects in EL images holds significant importance in photovoltaics. This paper introduces a state-of-the-art defect detection model based on the Yolo v.7 architecture designed explicitly for photovoltaic cell electroluminescence images. The model is trained to recognize and categorize five common defect classes, namely black core (Bc), crack (Ck), finger (Fr), star crack (Sc), and thick line (Tl). The proposed model exhibits remarkable performance through experimentation with an average precision of 80%, recall of 87%, and an mAP@.5 score of 86% across all defect classes. Furthermore, a comparative analysis is conducted to evaluate the model’s performance against two recently proposed models. The results affirm the excellent performance of the proposed model, highlighting its superiority in defect detection within the context of photovoltaic cell electroluminescence images. © 2024, The Author(s), under exclusive license to Springer Nature Switzerland AG.Öğe Short Message Service Spam Detection System for Securing Mobile Text Communication Based on Machine Learning(Springer Science and Business Media Deutschland GmbH, 2024) Malik, A.; Parihar, V.; Bhushan, B.; Hameed, A.A.; Jamil, A.; Bhattacharya, P.In recent years, the popularity of Short Message Service (SMS) on mobile phones has surged due to technological advancements and the growing prevalence of content-based advertising. However, this has also led to a surge in spam SMS, which can inundate one’s handset at any time and potentially result in the theft of personal information. Researchers have explored a range of options to combat spam SMS, including content-based machine learning algorithms and stylometric approaches. While filtering spam emails has proven to be effective, detecting spam SMS presents a unique set of challenges due to the presence of idioms, abbreviations, and well-known terms and phrases that are frequently used in legitimate messages. This study aims to examine and assess different classification techniques by utilizing datasets gathered from previous research studies. The primary emphasis is on comparing conventional Machine Learning (ML) methods. This study specifically investigates the efficacy of classification algorithms, including Logistic Regression (LR), Naïve Bayes (NB), and Random Forest (RF), in accurately identifying spam SMS messages. The results of our research demonstrate that the RF classifier outperforms other traditional ML techniques in the detection of spam SMS messages. This paper thoroughly examines diverse classification techniques employed in spam detection. The knowledge derived from this study has the potential to advance the development of more effective systems for detecting spam SMS messages. Ultimately, this would enhance the security and privacy of mobile phone users. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.Öğe A Survey on Image-Based Cardiac Diagnosis Prediction Using Machine Learning and Deep Learning Techniques(Springer Science and Business Media Deutschland GmbH, 2024) Nag, A.; Das, B.; Sil, R.; Hameed, A.A.; Bhushan, B.; Jamil, A.Cardiac imaging is crucial in the diagnosis of cardiovascular disease. Cardiovascular disease is the umbrella term for the majority of heart ailments. The majority of the causes of mortality are associated with cardiovascular illness. The authors provide a technique for the diagnosis of cardiac disease. The main aim of this study is to determine the most effective technique for predicting cardiovascular disease, specifically focusing on the use of signs of heart disease and Electrocardiogram images. This will be achieved by leveraging the latest advancements in Deep Learning and Machine Learning methods. The authors conduct a comprehensive examination of various Machine Learning and Deep Learning Techniques. These techniques were evaluated in the context of predicting cardiovascular disease evaluating Image. The analysis shows that the Convolutional Neural Network methods are much more effective than the alternatives. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.
