Automatic Image Segmentation for Lung using Deep Learning and Convolutional Neural Network

With advance in technology, rapidly growing medical treatment and healthcare which can cure or pre-detect the diagnosis. Lung segmentation (LS) is prerequisite step for lung image analysis to provide accurate lung image. Doctors usually detect diagnosis by checking X-ray which is very time consuming and tedious. Here, we demonstrate LS in using CXR images and evaluate which contents of the image influenced the most. Semantic segmentation (SS) was performed using a U-Net CNN architecture, and the classification using three CNN architectures. Segmentation with deep learning (DL) is having very similar accuracy as detecting diagnosis by doctors. Here, we demonstrate LS by using chest X-ray and segmentation was performed using U-net architecture. In this project we have connected this model which can easily separating Lung. The paper is detailed analysis and discussion of U-Net results and implementation of UNet in LS using X-ray