Scholarlink

A global pandemic has been ushered in by the recent coronavirus flare-up, which has thrown the whole globe into an unforeseen and disastrous predicament, endangering countless lives. It is critical to identify affected persons by rapid symptomatic contact in order to prevent the local spread of this virus. Chest X-ray (CXR) imaging, on the other hand, is more practical than computed tomography (CT) imaging due to its lower cost, simpler operation, and higher imaging speed. The reason for this research is to offer a method for classification of the Coronavirus via the use of CXR images and image processing. The image is resized, grayscale conversion, noise removal, thresholding, and morphological operation are used for preprocessing, train-test splitting is used for expressing the features, and categorization is done using multiple machine learning (ML) classifiers like Naive Bayes (NB), Decision Tree (DT), Random Forest (RF), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Bagging, AdaBoost, Gradient Boosting (GBoost) and Logistic Regression (LR). With bagging approach, the evaluation of results reveals a greater level of exactness, which is 95.19%. In order to evaluate the trial outcomes, execution metrics that were generated by the recommended approach are used as the basis for assessment.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

COVID-19 Forecasting Using CXR Images and Machine Learning Methods

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Shreeharsha Dash,
  • Subhalaxmi Das

摘要

A global pandemic has been ushered in by the recent coronavirus flare-up, which has thrown the whole globe into an unforeseen and disastrous predicament, endangering countless lives. It is critical to identify affected persons by rapid symptomatic contact in order to prevent the local spread of this virus. Chest X-ray (CXR) imaging, on the other hand, is more practical than computed tomography (CT) imaging due to its lower cost, simpler operation, and higher imaging speed. The reason for this research is to offer a method for classification of the Coronavirus via the use of CXR images and image processing. The image is resized, grayscale conversion, noise removal, thresholding, and morphological operation are used for preprocessing, train-test splitting is used for expressing the features, and categorization is done using multiple machine learning (ML) classifiers like Naive Bayes (NB), Decision Tree (DT), Random Forest (RF), K-Nearest Neighbors (KNN), Support Vector Machine (SVM), Bagging, AdaBoost, Gradient Boosting (GBoost) and Logistic Regression (LR). With bagging approach, the evaluation of results reveals a greater level of exactness, which is 95.19%. In order to evaluate the trial outcomes, execution metrics that were generated by the recommended approach are used as the basis for assessment.