A hybrid CNN and ensemble model for COVID-19 lung infection detection on chest CT scans

Abstract

COVID-19 is highly infectious and causes acute respiratory disease. Machine learning (ML) and deep learning (DL) models are vital in detecting disease from computerized chest tomography (CT) scans. The DL models outperformed the ML models. For COVID-19 detection from CT scan images, DL models are used as end-to-end models. Thus, the performance of the model is evaluated for the quality of the extracted feature and classification accuracy. There are four contributions included in this work. First, this research is motivated by studying the quality of the extracted feature from the DL by feeding these extracted to an ML model. In other words, we proposed comparing the end-to-end DL model performance against the approach of using DL for feature extraction and ML for the classification of COVID-19 CT scan images. Second, we proposed studying the effect of fusing extracted features from image descriptors, e.g., Scale-Invariant Feature Transform (SIFT), with extracted features from DL models. Third, we proposed a new Convolutional Neural Network (CNN) to be trained from scratch and then compared to the deep transfer learning on the same classification problem. Finally, we studied the performance gap between classic ML models against ensemble learning models. The proposed framework is evaluated using a CT dataset, where the obtained results are evaluated using five different metrics The obtained results revealed that using the proposed CNN model is better than using the well-known DL model for the purpose of feature extraction. Moreover, using a DL model for feature extraction and an ML model for the classification task achieved better results in comparison to using an end-to-end DL model for detecting COVID-19 CT scan images. Of note, the accuracy rate of the former method improved by using ensemble learning models instead of the classic ML models. The proposed method achieved the best accuracy rate of 99.39%.

A brief review and scientometric analysis on ensemble learning methods for handling COVID-19

“… State-of-the-art performance: CNNs have demonstrated excellent performance in various image classification tasks, surpassing human-level performance in some cases. Their ability to learn complex representations and generalize well from large datasets has made them a popular choice for COVID-19 detection, where an accurate and reliable diagnosis is crucial [ 99 , 100 ]. …”

Fighting against COVID‐19: Innovations and applications

“…In patients with COVID‐19, chest CT can reveal characteristic patterns of lung involvement, such as ground‐glass opacities, consolidation, and crazy‐paving patterns 30 . These findings can aid in diagnosing COVID‐19 and help healthcare professionals differentiate it from other respiratory conditions 31 …”

“…30 These findings can aid in diagnosing COVID-19 and help healthcare professionals differentiate it from other respiratory conditions. 31 Chest CT can also be used to monitor the progression of lung damage in COVID-19 patients, which can help healthcare professionals determine the appropriate treatment course. 32 For example, if the lung damage worsens despite treatment, a more aggressive treatment approach may be necessary.…”

Classification of Thoracic X-Ray Images of COVID-19 Patients Using the Convolutional Neutral Network (CNN) Method

“…They claim that the 3D CNN model is able to produce better performance than the 2D model in identifying signs of COVID-19 infection (Kumar et al, 2023). Although in almost all countries the status of COVID-19 from a pandemic has dropped to an endemic, research related to the detection of COVID-19 or abnormalities in the lungs using CNN models is still being carried out by many researchers (Kaya and Gürsoy, 2023;Gulakala et al, 2023;Kanjanasurat et al, 2023;Akl et al, 2023;Hussein et al, 2023;Sejuti and Islam, 2023;Liang et al, 2021;Oktamuliani et al, 2023) This article reports the development of COVID-19 detection using CNN and Roboflow, the application of the pseudocolour method was carried out in this study. The use of this method is used to improve visualization or contrast in X-ray chest images.…”