U-NET: A NOVEL METHOD FOR LUNG SEGMENTATION OF CHEST WITH CONVOLUTIONAL NEURAL NETWORK
No Thumbnail Available
Date
2019
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
UMT, Lahore
Abstract
Tuberculosis is a major health threat in many regions of the world. Opportunistic infections in immunocompromised HIV/AIDS patients and multi-drug-resistant bacterial strains have exacerbated the problem, while diagnosing tuberculosis still remains a challenge. Medical images have made a great impact on medicine, diagnosis, and treatment. The most important part of image processing is image segmentation. In this thesis I have described the latest segmentation methods applied in medical image analysis, I propose a novel method of X-ray of lungs segmentation using U-Net model. I construct the U-net which combine the lungs and mask. Then I convert to problem of positive and negative TB lungs into the segmentation of lungs, and extract the lungs by subtracting the chest from the radiography. In experiment, our model achieve 97.62% on the public dataset of collection of by Shenzhen Hospital, China and Montgomery County X-ray Set.