: The electrocardiogram (ECG) is a vital tool for detecting heart abnormalities, However, noise frequently disrupts the signals during recording, reducing diagnostic precision. During wireless recording and portable heart monitoring, one major source of noise is called additive white Gaussian noise (AWGN). Therefore, clean ECG signals are really important to diagnose cardic disorders. To address this concern , a novel approach is introduced that employs the Wavelet Packet Transform (WPT) for effective ECG signal denoising. WPT provides a comprehensive signal analysis, using the Symlets 8 mother wavelet function, decomposing ECG data into high and low frequency components over two levels. Subsequent to this, a soft thresholding (ST) technique is implemented to attenuate noise. Moreover, the universal threshold technique is incorporated, dynamically determining threshold values. Proposed method efficiently reduces noise through thresholding, addressing both low and high frequency noise components at each level. The retained coefficients are then utilized in the inverse WPT to reconstruct the denoised ECG signal. Comprehensive analysis highlights the robustness of our approach, demonstrating better performance compared to established denoising techniques on the MIT-BIH database. Performance metrics including Signal-to-Noise Ratio (SNR), SNR Improvement (SNRimp), correlation coefficient (CC) , Percentage Root Mean Square Difference (PRD) and Mean Squared Error (MSE) are employed. Proposed WPT approach, tailored through suitable decomposition levels and mother wavelet selection, represents a substantial improvement in ECG signal denoising beyond conventional techniques. The proposed method showcases substantial improvements over EMD-DWT, with 28.32% lower RMSE, 34.99% higher SNR, and 0.25% enhanced CC
{"title":"A Novel Approach for Denoising ECG Signals Corrupted with\u0000White Gaussian Noise Using Wavelet Packet Transform and\u0000Soft-Thresholding","authors":"Haroon Yousuf Mir, Omkar Singh","doi":"10.12785/ijcds/150196","DOIUrl":"https://doi.org/10.12785/ijcds/150196","url":null,"abstract":": The electrocardiogram (ECG) is a vital tool for detecting heart abnormalities, However, noise frequently disrupts the signals during recording, reducing diagnostic precision. During wireless recording and portable heart monitoring, one major source of noise is called additive white Gaussian noise (AWGN). Therefore, clean ECG signals are really important to diagnose cardic disorders. To address this concern , a novel approach is introduced that employs the Wavelet Packet Transform (WPT) for effective ECG signal denoising. WPT provides a comprehensive signal analysis, using the Symlets 8 mother wavelet function, decomposing ECG data into high and low frequency components over two levels. Subsequent to this, a soft thresholding (ST) technique is implemented to attenuate noise. Moreover, the universal threshold technique is incorporated, dynamically determining threshold values. Proposed method efficiently reduces noise through thresholding, addressing both low and high frequency noise components at each level. The retained coefficients are then utilized in the inverse WPT to reconstruct the denoised ECG signal. Comprehensive analysis highlights the robustness of our approach, demonstrating better performance compared to established denoising techniques on the MIT-BIH database. Performance metrics including Signal-to-Noise Ratio (SNR), SNR Improvement (SNRimp), correlation coefficient (CC) , Percentage Root Mean Square Difference (PRD) and Mean Squared Error (MSE) are employed. Proposed WPT approach, tailored through suitable decomposition levels and mother wavelet selection, represents a substantial improvement in ECG signal denoising beyond conventional techniques. The proposed method showcases substantial improvements over EMD-DWT, with 28.32% lower RMSE, 34.99% higher SNR, and 0.25% enhanced CC","PeriodicalId":37180,"journal":{"name":"International Journal of Computing and Digital Systems","volume":"51 4","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140254818","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}
Atrab A. Abd El-Aziz, Reda A. El-Khoribi, Nour Eldeen Khalifa
: Attacks against deep learning (DL) models are considered a significant security threat. However, DL especially deep convolutional neural networks (CNN) has shown extraordinary success in a wide range of medical applications, recent studies have recently proved that they are vulnerable to adversarial attacks. Adversarial attacks are techniques that add small, crafted perturbations to the input images that are practically imperceptible from the original but misclassified by the network. To address these threats, in this paper, a novel defense technique against white-box adversarial attacks based on CNN fine-tuning using the weights of the pre-trained deep convolutional autoencoder (DCAE) called Robust Defense Model against Adversarial Attacks (RDMAA), for DL-based cancer diagnosis is introduced. Before feeding the classifier with adversarial examples, the RDMAA model is trained where the perpetuated input samples are reconstructed. Then, the weights of the previously trained RDMAA are used to fine-tune the CNN-based cancer diagnosis models. The fast gradient method (FGSM) and the project gradient descent (PGD) attacks are applied against three DL-cancer modalities (lung nodule X-ray, leukemia microscopic, and brain tumor magnetic resonance imaging (MRI)) for binary and multiclass labels. The experiment’s results proved that under attacks, the accuracy decreased to 35% and 40% for X-rays, 36% and 66% for microscopic, and 70% and 77% for MRI. In contrast, RDMAA exhibited substantial improvement, achieving a maximum absolute increase of 88% and 83% for X-rays, 89% and 87% for microscopic cases, and 93% for brain MRI. The RDMAA model is compared with another common technique (adversarial training) and outperforms it. Results show that DL-based cancer diagnoses are extremely vulnerable to adversarial attacks, even imperceptible perturbations are enough to fool the model. The proposed model RDMAA provides a solid foundation for developing more robust and accurate medical DL models.
{"title":"RDMAA: Robust Defense Model against Adversarial Attacks\u0000in Deep Learning for Cancer Diagnosis","authors":"Atrab A. Abd El-Aziz, Reda A. El-Khoribi, Nour Eldeen Khalifa","doi":"10.12785/ijcds/150190","DOIUrl":"https://doi.org/10.12785/ijcds/150190","url":null,"abstract":": Attacks against deep learning (DL) models are considered a significant security threat. However, DL especially deep convolutional neural networks (CNN) has shown extraordinary success in a wide range of medical applications, recent studies have recently proved that they are vulnerable to adversarial attacks. Adversarial attacks are techniques that add small, crafted perturbations to the input images that are practically imperceptible from the original but misclassified by the network. To address these threats, in this paper, a novel defense technique against white-box adversarial attacks based on CNN fine-tuning using the weights of the pre-trained deep convolutional autoencoder (DCAE) called Robust Defense Model against Adversarial Attacks (RDMAA), for DL-based cancer diagnosis is introduced. Before feeding the classifier with adversarial examples, the RDMAA model is trained where the perpetuated input samples are reconstructed. Then, the weights of the previously trained RDMAA are used to fine-tune the CNN-based cancer diagnosis models. The fast gradient method (FGSM) and the project gradient descent (PGD) attacks are applied against three DL-cancer modalities (lung nodule X-ray, leukemia microscopic, and brain tumor magnetic resonance imaging (MRI)) for binary and multiclass labels. The experiment’s results proved that under attacks, the accuracy decreased to 35% and 40% for X-rays, 36% and 66% for microscopic, and 70% and 77% for MRI. In contrast, RDMAA exhibited substantial improvement, achieving a maximum absolute increase of 88% and 83% for X-rays, 89% and 87% for microscopic cases, and 93% for brain MRI. The RDMAA model is compared with another common technique (adversarial training) and outperforms it. Results show that DL-based cancer diagnoses are extremely vulnerable to adversarial attacks, even imperceptible perturbations are enough to fool the model. The proposed model RDMAA provides a solid foundation for developing more robust and accurate medical DL models.","PeriodicalId":37180,"journal":{"name":"International Journal of Computing and Digital Systems","volume":"31 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140255246","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}
: This paper introduces an e ffi cient method for 3D instance segmentation based on 2D object detection, applied to the photogrammetric survey images of archaeological sites. The method capitalizes on the relationship between the 3D model and the set of 2D images utilized to compute it. 2D detections on the images are projected and transformed into a 3D instance segmentation, thus identifying unique objects within the scene. The primary contribution of this work is the development of a semi-automatic image annotation method, augmented by an object tracking technique that leverages the temporal continuity of image sequences. Additionally, a novel ad-hoc evaluation process has been integrated into the conventional annotation-training-testing cycle to determine the necessity of additional annotations. This process tests the consistency of the 3D objects yielded by the 2D detection. The e ffi cacy of the proposed method has been validated on the underwater site of Xlendi in Malta, resulting in complete and accurate 3D instance segmentation. Compared to traditional methods, the object tracking approach adopted has facilitated a 90% reduction in the need for manual annotations, The approach streamlines precise 3D detection, establishing a robust foundation for comprehensive 3D instance segmentation. This enhancement enriches the 3D survey, providing profound insights and facilitating seamless exploration of the Xlendi site from an archaeological perspective.
{"title":"Efficient 3D Instance Segmentation for Archaeological Sites\u0000Using 2D Object Detection and Tracking","authors":"Maad kamal Al-anni, Pierre Drap","doi":"10.12785/ijcds/150194","DOIUrl":"https://doi.org/10.12785/ijcds/150194","url":null,"abstract":": This paper introduces an e ffi cient method for 3D instance segmentation based on 2D object detection, applied to the photogrammetric survey images of archaeological sites. The method capitalizes on the relationship between the 3D model and the set of 2D images utilized to compute it. 2D detections on the images are projected and transformed into a 3D instance segmentation, thus identifying unique objects within the scene. The primary contribution of this work is the development of a semi-automatic image annotation method, augmented by an object tracking technique that leverages the temporal continuity of image sequences. Additionally, a novel ad-hoc evaluation process has been integrated into the conventional annotation-training-testing cycle to determine the necessity of additional annotations. This process tests the consistency of the 3D objects yielded by the 2D detection. The e ffi cacy of the proposed method has been validated on the underwater site of Xlendi in Malta, resulting in complete and accurate 3D instance segmentation. Compared to traditional methods, the object tracking approach adopted has facilitated a 90% reduction in the need for manual annotations, The approach streamlines precise 3D detection, establishing a robust foundation for comprehensive 3D instance segmentation. This enhancement enriches the 3D survey, providing profound insights and facilitating seamless exploration of the Xlendi site from an archaeological perspective.","PeriodicalId":37180,"journal":{"name":"International Journal of Computing and Digital Systems","volume":"19 21","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140396438","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}
Suprapto Suprapto, W. Wahyono, Nur Rokhman, Faisal Dharma Adhinata
: Machine learning is crucial in categorizing data into specific classes based on their features. However, challenges emerge, especially in classification, when dealing with imbalanced datasets. An imbalanced dataset occurs when there is a disproportionate number of samples across di ff erent classes. It leads to a machine learning model’s bias towards the majority class and poor recognition of minority classes, often resulting in notable prediction inaccuracies for those less represented classes. This research proposes a cascade and parallel architecture in the training process to enhance accuracy and speed compared to non-cascade and sequential. This research will evaluate the performance of the SVM and Random Forest methods. Our findings reveal that employing the Random Forest method, configured with 100 trees, substantially enhances classification accuracy by 4.72%, elevating it from 58.87% to 63.59% compared to non-cascade classifiers. Furthermore, adopting the Message Passing Interface for Python (MPI4Py) for parallel processing across multiple cores or nodes demonstrates a remarkable increase in training speed. Specifically, parallel processing was found to accelerate the training process by up to 4.35 times, reducing the duration from 1725.86 milliseconds to a mere 396.54 milliseconds. These results highlight the advantages of integrating parallel processing with a cascade architecture in machine learning models, particularly in addressing the challenges associated with imbalanced datasets. This research demonstrates the potential for substantial improvements in classification tasks’
{"title":"A Parallel Approach of Cascade Modelling Using MPI4Py on\u0000Imbalanced Dataset","authors":"Suprapto Suprapto, W. Wahyono, Nur Rokhman, Faisal Dharma Adhinata","doi":"10.12785/ijcds/150191","DOIUrl":"https://doi.org/10.12785/ijcds/150191","url":null,"abstract":": Machine learning is crucial in categorizing data into specific classes based on their features. However, challenges emerge, especially in classification, when dealing with imbalanced datasets. An imbalanced dataset occurs when there is a disproportionate number of samples across di ff erent classes. It leads to a machine learning model’s bias towards the majority class and poor recognition of minority classes, often resulting in notable prediction inaccuracies for those less represented classes. This research proposes a cascade and parallel architecture in the training process to enhance accuracy and speed compared to non-cascade and sequential. This research will evaluate the performance of the SVM and Random Forest methods. Our findings reveal that employing the Random Forest method, configured with 100 trees, substantially enhances classification accuracy by 4.72%, elevating it from 58.87% to 63.59% compared to non-cascade classifiers. Furthermore, adopting the Message Passing Interface for Python (MPI4Py) for parallel processing across multiple cores or nodes demonstrates a remarkable increase in training speed. Specifically, parallel processing was found to accelerate the training process by up to 4.35 times, reducing the duration from 1725.86 milliseconds to a mere 396.54 milliseconds. These results highlight the advantages of integrating parallel processing with a cascade architecture in machine learning models, particularly in addressing the challenges associated with imbalanced datasets. This research demonstrates the potential for substantial improvements in classification tasks’","PeriodicalId":37180,"journal":{"name":"International Journal of Computing and Digital Systems","volume":"63 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140254789","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}
{"title":"A Comprehensive Comparative Study of Machine Learning Algorithms for Water Potability Classification","authors":"Fuad Ahmad Musleh","doi":"10.12785/ijcds/150184","DOIUrl":"https://doi.org/10.12785/ijcds/150184","url":null,"abstract":"","PeriodicalId":37180,"journal":{"name":"International Journal of Computing and Digital Systems","volume":"15 5","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140084653","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}
{"title":"Implementing Image Processing and Deep Learning\u0000Techniques to Analyze Skin Cancer Images","authors":"Snowber Mushtaq, Omkar Singh","doi":"10.12785/ijcds/150188","DOIUrl":"https://doi.org/10.12785/ijcds/150188","url":null,"abstract":"","PeriodicalId":37180,"journal":{"name":"International Journal of Computing and Digital Systems","volume":"33 16","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140082534","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}
{"title":"Studying Vibratory Patterns of Vocal Folds and Their\u0000Impairments in Parkinson’s Disease: A Theoretical Approach","authors":"R. Indu, Sushil Chandra Dimri","doi":"10.12785/ijcds/150182","DOIUrl":"https://doi.org/10.12785/ijcds/150182","url":null,"abstract":"","PeriodicalId":37180,"journal":{"name":"International Journal of Computing and Digital Systems","volume":"32 9","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140084199","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}
Sanjeev Dhawan, Kulvinder Singh, Amit Batra, Anthony Choi, Ethan Choi
{"title":"An Optimized Ranking Based Technique towards\u0000Conversational Recommendation Models","authors":"Sanjeev Dhawan, Kulvinder Singh, Amit Batra, Anthony Choi, Ethan Choi","doi":"10.12785/ijcds/150185","DOIUrl":"https://doi.org/10.12785/ijcds/150185","url":null,"abstract":"","PeriodicalId":37180,"journal":{"name":"International Journal of Computing and Digital Systems","volume":"34 12","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140087490","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}
Abderrahim El Mhouti, Mohamed Fahim, Asmae Bahbah, Yassine El Borji, Adil Soufi, M. Erradi
{"title":"Using Cloud Services to Improve Weather Forecasting Based\u0000on Weather Big Data Scraped From Web Sources","authors":"Abderrahim El Mhouti, Mohamed Fahim, Asmae Bahbah, Yassine El Borji, Adil Soufi, M. Erradi","doi":"10.12785/ijcds/150183","DOIUrl":"https://doi.org/10.12785/ijcds/150183","url":null,"abstract":"","PeriodicalId":37180,"journal":{"name":"International Journal of Computing and Digital Systems","volume":" 927","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140092031","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}
Moayad Almutairi, Shailen Mishra, Mohammed AlShehri
{"title":"Forensics Analysis of Cloud-Computing Traffics","authors":"Moayad Almutairi, Shailen Mishra, Mohammed AlShehri","doi":"10.12785/ijcds/150173","DOIUrl":"https://doi.org/10.12785/ijcds/150173","url":null,"abstract":"","PeriodicalId":37180,"journal":{"name":"International Journal of Computing and Digital Systems","volume":"104 34","pages":""},"PeriodicalIF":0.0,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140089995","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：