CNN-Based Covid-19 Detection from Two Distinct Chest X-Ray Datasets: Leveraging TensorFlow and Keras for Novel Results

The Covid-19 pandemic has profoundly influenced global health and daily life across numerous countries, necessitating the urgent implementation of effective diagnostic strategies. This underscores the importance of advancing accurate, efficient, and rapid early detection techniques. In this context, convolutional neural networks (CNNs) have demonstrated remarkable proficiency in image recognition and classification tasks, particularly when applied to large annotated datasets. However, the domain of medical image classification presents significant challenges primarily stemming from the scarcity of annotated medical images such as chest X-rays images. Therefore, this study presents a new deep learning model for Covid-19 diagnosis from chest X-rays. Two distinct chest X-ray datasets from different sources are utilized for model training and testing. The proposed CNN-based model accurately calculates chest X-rays into positive and negative categories, providing an automated and efficient approach to diagnosing viral disease. This work holds significant importance for pandemic control and a safer future.