Automatic detection of pulmonary tuberculosis using image processing techniques

Tuberculosis is a major problem and rapidly spread disease in all over the world. Accurate diagnosis is the key to controlling the disease. Traditional methods like tuberculin skin test (TST), Acid fast staining produce results that are inaccurate or take more time to detect. This paper presents an automated approach to detect tuberculosis using chest radiographs. Chest radiographic images is chosen to detect tuberculosis. In the existing method, cavity detection, ribs and diaphragm elimination is difficult to examine tuberculosis in chest radiographs. To overcome the difficulties lung region is extracted by using registration based segmentation methods. Segmentation of lung regions is performed after the registration process to handle complex segmentation problems. The performance of our system is evaluated by using two datasets: Montgomery country (MC) and Japanese society of radiology(JSRT) dataset and compare the results with the existing method to determine the accurate results.