Prediction of Lung Diseases Using Deep Learning Models

Recent developments in deep learning support have led to the identification and classification of lung disease in medical images. Hence, there are many studies on lung disease detection using deep learning, which can be found in the literature. The development of precise and effective predictive models for early identification and risk assessment is necessary due to the increasing incidence of lung illnesses. An area of artificial intelligence called deep learning has demonstrated potential in a number of medical applications, including risk assessment. Utilizing a large dataset that includes patient demographic data, clinical data, and medical imaging data, this work focused on applying deep learning algorithms to predict the risk of lung diseases. Cleaning, normalizing, and adding features to improve the robustness of the model are all parts of the data preprocessing step. The best risk prediction is achieved by utilizing a well-selected deep learning architecture that combines convolutional neural networks (CNNs) for image data with conventional neural networks for tabular data. The lung X-ray images are cast off to train the model, and every piece of data is carefully examined.