Early Detection of Thoracic Disease using Segmentation based Deep Convolutional Neural Network

The area of medical imaging has benefited greatly from deep learning. Thoracic issues can be fatal if they are not promptly identified and treated since they can be quite significant. Numerous thoracic illnesses can be diagnosed using chest X-ray imaging. We frequently needed a qualified and experienced radiologist to assess chest X-ray images and determine whether or not a person has certain thorax abnormalities. Because prompt diagnoses are sometimes necessary, not everyone has the chance to see an expert about treating chest disease in a timely manner. In order to help clinicians make important decisions, we outline a technique that uses picture classification and segmentation to detect a specific thoracic condition and produce a mask. Deep Learning techniques have outperformed several state-of-the-art models in a variety of scenarios, demonstrating their value. With ResNet serving as the foundation, we achieved successful outcomes using the U-Net architecture. U-Net performs better in the areas of semantic segmentation and medical image processing