Covid-19 Hastalığının Teşhisi için CNN Tabanlı Modeller ile Adaboost Algoritmasının Kombinasyonunun Performans Analizi

Abstract

At the end of 2019, Covid-19, which is a new form of Coronavirus, has spread widely all over the world. With the increasing daily cases of this disease, fast, reliable, and automatic detection systems have been more crucial. Therefore, this study proposes a new technique that combines the machine learning algorithm of Adaboost with Convolutional Neural Networks (CNN) to classify Chest X-Ray images. Basic CNN algorithm and pretrained ResNet-152 have been used separately to obtain features of the Adaboost algorithm from Chest X-Ray images. Several learning rates and the number of estimators have been used to compare these two different feature extraction methods on the Adaboost algorithm. These techniques have been applied to the dataset, which contains Chest X-Ray images labeled as Normal, Viral Pneumonia, and Covid-19. Since the used dataset is unbalanced between classes SMOTE method has been used to make the number of images of classes balance. This study shows that proposed CNN as a feature extractor on the Adaboost algorithm(learning rate of 0.1 and 25 estimators) provides higher classification performance with 94.5% accuracy, 93% precision, 94% recall, and 93% F1-score.

Predicting ICU Admission for COVID-19 Patients in Saudi Arabia: A Comparative Study of AdaBoost and Bagging Methods

“…Darici [15] performed a comparative analysis between the AdaBoost-CNN and AdaBoost-ResNet-152 methods to not only autonomously extract image features from X-ray chest COVID-19 patients but also classify these images. The authors used datasets containing 2,905 photos from various sources, with an unequal distribution across classes, and the SMOTE was used to balance the number of photos in each class.…”

“…As an outcome of Saudi nationwide quantitative study of RT-PCR 15 tests, a private COVID-19 Positive Patients (CDPP) dataset was aggregated. The CDPP dataset was curated under several Saudi authorities, including the Global Center for Mass Gatherings Medicine, the Saudi National Health Laboratory, and the Saudi Health Electronic Surveillance Network 16 [21,22].…”

“…The dataset comprised 639 records, with 44 features that 15 Reverse Transcription Polymerase Chain Reaction 16 Site: https://hesn.moh.gov.sa/webportal/ included clinical and demographic information about symptomatic and asymptomatic patients. There are three types of variables in the dataset: Boolean (15), categorical (15), and numerical (18). Table I To visualize the distribution and intensity of data points in the CDPP dataset, Fig.…”

A Deep Learning Based on Automatic Cerebral Aneurysm Detection in Brain Computed Tomography Angiography Scan Images

“…In addition, aneurysm detection was performed by evaluating DSA images in the ready dataset obtained from the Cerebral Aneurysm Detection and Analysis (CADA) competition [18]. Other studies have similar content [20][21][22][23]. The motivation of the study, a new dataset that is not available in the literature was used.…”

Dezenformasyonun Otomatik Tespiti: Sistematik Bir Haritalama Çalışması

“…In addition, the fact that the studies carried out belong to authors from many different countries shows that there is a worldwide interest in the research field and that it has great potential. Also, it is stated in the literature that with the increasing number of developing technologies, imaging techniques and open access data sets, methods such as artificial intelligence, machine learning and deep learning have become popular and are widely used in many fields [44]. It was observed that the most frequently used keywords by the authors of the studies were: fake news detection, machine learning, deep learning, fake news, convolutional neural networks (CNN) and COVID-19.…”