Multi-Classification of Chest X-rays for COVID-19 Diagnosis Using Deep Learning Algorithms

AbdElhamid, Abeer A.; AbdElhalim, Eman; Mohamed, Mohamed A.; Khalifa, Fahmi

doi:10.3390/app12042080

Cited by 18 publications

(8 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In [34], the authors developed the global average pooling-based CNN (GAP-CNN) model for classifying three COVID-19 classes. However, this method did not implement any segmentation operations.…”

Section: B Survey On Covid-19 Disease Diagnosis Modelsmentioning

confidence: 99%

Securing IoMT: A Hybrid Model for DDoS Attack Detection and COVID-19 Classification

Sripriyanka,

Mahendran

2024

IEEE Access

View full text Add to dashboard Cite

The coronavirus disease 2019 (COVID-19) epidemic had a momentous influence on the state of universal health and how people live their lives in every nation. The combination of the Internet of Things (IoT) with medical systems is referred to as the "Internet of Medical Things (IoMT)," and it makes it possible for various medical events to take place, including real-time medicine prescriptions, remote patient observing, and real-time diagnosis, amongst other things. However, the security, integrity, and concealment of medical data on the IoMT remain an endless issue, contributing to the difficulties that arise in providing medical services. However, these conventional methods failed to provide robust security against attacks like distributed denial of service (DDoS). As a result, security IoMT-based COVID-19 detection applications are required. Therefore, this article proposes a novel IoMT-based COVID-19 detection and classification network (ICDC-Net) for smart healthcare applications. Initially, Optimal Feistel Block Cipher (OFBC) encryption was performed on COVID-19-based medical data (chest x-ray images), which provides the highest security to the patient's data. Here, OFBC loss is optimized by the Hybrid Grey-Wolf Optimizer with Particle Swarm Optimization (HGWO-PSO), which performs DDoS attack detection and prevention. Further, the HGWO-PSO is also used to extract disease-specific features from the COVID-19 chest X-ray (CXR) images. Finally, the Residual Network50 (ResNet50) deep learning model is used to classify multiple diseases from CXR images, including normal, COVID-19, and viral pneumonia. The simulations disclosed that the proposed ICDC-Net resulted in improved attack detection accuracy (ADA), attack detection time (ADT), and reduced attack detection error rates (ADER) as compared to other security standards. Further, the simulations also disclosed that the proposed ICDC-Net developed superior COVID-19 categorization performing as equated to existing classification prototypes.

show abstract

Section: B Survey On Covid-19 Disease Diagnosis Modelsmentioning

confidence: 99%

Securing IoMT: A Hybrid Model for DDoS Attack Detection and COVID-19 Classification

Sripriyanka,

Mahendran

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…According to the WHO, long-term COVID-19 is diagnosed several days after the infection is confirmed. It said longterm COVID-19 effects can be seen for at least 90 days after the initial symptoms clear up [4].…”

Section: Related Workmentioning

confidence: 99%

“…However, some patients with neurological disabilities experience various lingering symptoms even months after the infection [2]. According to the World Health Organization (WHO), it has been mentioned that some long-term health disorders may occur in people who have been exposed to the coronavirus for a long time [3,4].…”

Section: Introductionmentioning

confidence: 99%

Examination of the Effects of Long-term COVID-19 Impacts on Patients with Neurological Disabilities Using a Neuromachine Learning Model

Vaniprabha¹,

Logeshwaran²,

Kiruthiga³

et al. 2023

BIJNN

View full text Add to dashboard Cite

Currently, studies have shown that one in three people infected with coronavirus disease-19 (COVID-19) is likely to have had long-term exposure to COVID-19, known as long-term COVID-19. Clinical studies indicate that many people infected with the severe acute respiratory syndrome Coronavirus-2 (SARS-CoV-2) COVID-19 pandemic have long-term COVID-19 exposure. According to the study, it has been said that people with diabetes and obesity, and people who have received organ transplants, are more likely to suffer from this long-term effect of COVID-19. In this article, the effects of long-term COVID-19 exposure on neurological disability patients are analyzed with the help of a neuromachine learning model. The proposed model also shows that this long-term COVID problem does not depend on the factors such as race, age, gender, and socioeconomic status of those people. According to the proposed model, people suffering from long-term COVID problems continue to suffer from physical fatigue and shortness of breath and are regularly monitored and classified as per the proposed instructions. Even after they recover from the disease, various side effects are seen.

show abstract

“…It consists of two main steps: first, training model on the dataset, and then the best models are aggregated by combining the predictions of the various best models. AbdElhamid et al [25] proposed a transfer learningbased multi-level diagnostic framework for detecting Covid-19 in X-ray images. The framework comprises three stages: pre-processing, feature extraction, and classification.…”

Section: Literature Reviewmentioning

confidence: 99%

An Automated Chest X-Ray Image Analysis for Covid-19 and Pneumonia Diagnosis Using Deep Ensemble Strategy

Hussain,

Amin,

Lee

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Precise and timely diagnosis of Covid-19 and pneumonia is crucial for effective treatment. However, the traditional RT-PCR method is time-consuming, costly, and prone to incorrect results. To address these limitations, a deep ensemble strategy is proposed as a promising alternative to provide more accurate and reliable outcomes. The strategy comprises three main stages: i) pre-processing, ii) salient feature extraction, and iii) training and classification. In the pre-processing step, the authors resize the images to the desired input shape. Data augmentation techniques, such as zooming, nearest full mode, rotation, and flipping, are employed to augment the dataset, thereby improving the training accuracy of the proposed approach. Additionally, the proposed method leverages the capabilities of VGG-16, DenseNet-201, and Efficient-B0 models using transfer-learning techniques to extract deep features from the images. These salient features are then passed through proposed fully connected layers and ensemble classifiers to predict the probability of the given classes. Extensive experiments were conducted on a chest X-ray image dataset, demonstrating that the proposed system outperforms contemporary techniques in terms of precision, recall, F1-score, and accuracy (acc). The proposed method obtained 97% of acc, while 96%, 95%, and 97% pre, rec, and F1-score respectively. In conclusion, this study presents a valuable contribution to medical image diagnosis using an AI-based deep ensemble strategy. The proposed approach offers a promising solution for accurate and efficient diagnosis of Covid-19 and pneumonia, assisting healthcare professionals in making informed decisions for optimal treatment outcomes.INDEX TERMS Chest x-ray images, covid-19 and pneumonia diagnosis, ensemble learning, fine-tuning, pattern recognition, transfer learning.

show abstract

Multi-Classification of Chest X-rays for COVID-19 Diagnosis Using Deep Learning Algorithms

Cited by 18 publications

References 41 publications

Securing IoMT: A Hybrid Model for DDoS Attack Detection and COVID-19 Classification

Securing IoMT: A Hybrid Model for DDoS Attack Detection and COVID-19 Classification

Examination of the Effects of Long-term COVID-19 Impacts on Patients with Neurological Disabilities Using a Neuromachine Learning Model

An Automated Chest X-Ray Image Analysis for Covid-19 and Pneumonia Diagnosis Using Deep Ensemble Strategy

Contact Info

Product

Resources

About