AI-Based Deep Random Forest Ensemble Model for Prediction of COVID-19 and Pneumonia from Chest X-Ray Images

• Published in: Artificial Intelligence for Innovative Healthcare Informatics pp. 133 - 149
• Main Authors: Jothi Prabha, A., Venkateswaran, N., Sengodan, Prabaharan
• Format: Book Chapter
• Language: English
• Published: Cham Springer International Publishing
• Subjects: Covid-19; Deep learning; Pneumonia; Random forest medical images
• ISBN: 9783030965686; 3030965686
• DOI: 10.1007/978-3-030-96569-3_7

The novel coronavirus commonly known as COVID-19 has spread rapidly with a high mortality rate all over the world. This pandemic has affected not only our health but also other sectors of our life including our lifestyle and global economy. There is a need for automated tools to detect covid-19 at an earlier stage to save lives, evade the swift spread of the disease, and reduce the mortality rate. Covid-19 mainly affects the respiratory system and its symptoms are usually cough, cold, fever, and difficulty in breathing. Severe Pneumonia has similar symptoms and it’s tough to discrete Covid-19 pneumonia and other kinds of Pneumonia. This research work proposes an AI-based deep random forest ensemble model that uses the key characteristics of deep learning and a multi-layer ensemble deep random forest model for early prediction of Covid-19 Pneumonia and other types of pneumonia. The proposed model has given better accuracy when compared to existing deep learning models. The major advantages of the proposed ensemble model are it does not need hyperparameter tuning and backpropagation hence the model is less complex. The results of the proposed AI-based deep random forest ensemble model are compared to the results of existing models. The detection rate of the proposed ensemble model is high (96.6%) compared to the state-of-art models. The proposed AI-based Deep Random Forest ensemble model has helped in the effective prediction of Covid-19 Pneumonia and other kinds of pneumonia based on Chest X-rays.