A Deep Learning Model for Early Prediction of Pneumonia Using VGG19 and Neural Networks

ABSTRACT

Pneumonia is a disease that can be caused by bacteria, viruses, and fungi. According to WHO, pneumonia is responsible for 22% of all deaths of children under the age of 1–5 years which is one of the main causes of increased mortality rate. Congestion, gray hepatization, red hepatization, and resolution are the stages of this disease. If the disease is not detected in time, it can progress to a fatal stage. The chest X-ray image is used to diagnose pneumonia, but it requires the presence of experienced radiologists. Pneumonia, COVID-19, cancer, and various other diseases can be identified using X-ray images. If the disease is incorrectly identified, severe difficulties may arise. A deep learning-based model called VGG19 is used to address this issue, which classifies pneumonia from normal lungs. A chest X-ray dataset containing 5856 images was used in this study to classify pneumonia from normal lungs. The outcomes have been demonstrated as accuracy, precision, recall, F1-score, and receiver operating characteristics with the values of 93%, 0.931, 0.93, 0.931, and 0.973, respectively. Furthermore, for validating the proposed model, the performance parameters are compared to the existing work, which results that the proposed model outperforms the other models. In future, this work could be used in hospitals and medical applications. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023.