Deep Learning based Early-Stage Restrictive Lung Disease Prediction
Abstract
The global healthcare system is severely burdened by lung diseases, which include a wide range of conditions like asthma, lung cancer,chronic obstructive pulmonary disease (COPD) and pulmonary fibrosis. This study examines the etiology, various diagnostic techniques, and cutting-edge therapies for common lung conditions that impact a sizable patient population.Through a critical systematic review of recent literature and clinical evidence, we conduct an extensive analysis of various risk factors such as but not limited to smoking, air pollution, and genetic predisposition that can cause these disorders. To overcome such difficulties, this research proposes a research method based on deep learning technology to effectively classify different lung diseases using CT scan images and chest X-ray images. Hence the superior Convolutional Neural Networks (CNNs), specifically employing models such as ResNet50, DenseNet performing well with higher accuracy for early prediction.