AUTOMATED FEATURE EXTRACTION AND CLASSIFICATION OF COVID-19 CHEST X-RAY IMAGES USING CONVOLUTIONAL NEURAL NETWORKS AND XGBOOST ALGORITHMS

Abstract

: In the upgrowing astounding development of the medical epoch, people are trying and fighting several diseases to extend their life. Ever since the arrival of covid –19 (SARS-CoV2) epidemic health crisis took over the world, was a major challenge to the global healthcare centre, to prevent human efforts of disease containment from being overwhelmed, we need tools that can streamline the diagnosis, surveillance, and treatment at right time and manage it properly. the detection of COVID-19 using the conventional method is time-consuming, not very accurate, and error-prone. During this period, the alternate solution of the medical X-ray image plays an important role in diagnosing COVID-19 patients effectively. This current research work takes the dataset of COVID-19-affected chest X-rays and proposes a Deep Learning (DL)-based approach. The images are pre-processed using Median Filter (MF) and Contrast-Limited Adaptive Histogram Equalization (CLAHE). The features are extracted using the Gray Level Co-occurrence Matrix (GLCM) and Convolutional Neural Network (CNN). The images are classified using Random Forest (RF) and eXtreme Gradient Boosting (XGBoost). Experimental results show the efficiency of the proposed model CNN_XGBoost offers better results in terms of Accuracy, Precision, Recall and F-Measure in contrast to RF, CNN and CNN_RF.