Detection of pulmonary nodules in CT images using template matching and neural classifier

Computer aided pulmonary nodule detection has been among major researchtopics lately to help for early treatment of lung cancer which is the most lethal kindof cancer worldwide.Some evidence suggests that periodic screening tests with theCT of patients will help in reducing the mortality rate caused by the lung cancer.Acomplete and accurate computer aided diagnosis (CAD) system for detection ofnodules in lung CT images consists of three main steps: extraction of lungparenchyma, candidate nodule detection and false positive reduction. While precisesegmentation of lung region speed upthe detection process of pulmonary nodules bylimiting the search area, in candidate nodule detection step we attempt to include allnodule like structures. However, the main problem in the current CAD systems fornodule detection is the high false positive rate which is mostly associated tomisrecognition of juxta-vascular nodules from blood vessels. In this paper wepropose an automated method which has all of the three above mentioned steps. Ourmethod attempts to find initial nodules by thresholding and template matching. Toseparate false positives from nodules we use feature extraction and neural classifier.The proposed method has been evaluated against several images in LIDC databaseand the results demonstrateimprovements comparing to previous methods.

Detection of pulmonary nodules in CT images using template matching and neural classifier Keywords:

False positive reduction , neural classifier , pulmonary nodule detection , template matching