Lung Disease Detection Based on Deep Learning Techniques: A Review

This survey paper intends to accumulate data to develop a system that assists in the early detection of abnormal lung conditions such as pneumonia, lung cancer, TB, and COVID-19. Early detection is the key factor for effective treatment and improving patient outcomes. To create a machine learning model capable of analyzing medical images (X-rays, CT scans) to identify abnormal lung patterns. This system will aid healthcare professionals in making faster and more accurate diagnoses. For the segmentation task, we have chosen the DRD U-Net algorithm through a rigorous evaluation of different parameters. This approach is designed to address the complex challenge of classifying abnormal lung images into multiple categories, effectively distinguishing between COVID-19, cancer, pneumonia, TB, and normal subjects. This model training will be an effective way of identifying and early detection of respiratory diseases. Early detection of abnormal lung conditions leads to timely treatment, improved patient care and cost savings for patients. For healthcare providers, this can be a valuable tool.