Feed forward Convolutional Neural Network Approach for Predicting Covid Disease

Abstract

Identifying students' progress at an early stage enables educators to refine teaching strategies and incorporate varied instructional methods to enhance the overall learning experience. Machine learning techniques offer valuable support by forecasting potential learning difficulties, allowing educators to intervene early and guide students more effectively. This study explores the use of classification techniques, specifically Decision Tree (DT) algorithms, to predict students' academic performance following their preparatory year and determine which algorithm delivers the highest accuracy. In parallel, the emergence of COVID-19 has led to significant mortality, particularly among the elderly and individuals with preexisting health conditions. The primary diagnostic approach for COVID-19 has been the reverse-transcription polymerase chain reaction (RT-PCR) test, which, although effective, is often costly and time-intensive. This has underscored the necessity for rapid, affordable diagnostic alternatives to aid clinical evaluations. To address this, we evaluated the performance of transfer learning by employing pretrained deep convolutional neural networks to detect COVID-19 using chest X-ray images, utilizing two publicly available datasets under various experimental conditions.