Lung cancer is remaining as a major cause of cancer death, where it makes up almost 25% of cancer deaths worldwide. As per the report of World Health Organization (WHO), around 18 million people die due to lung cancer every year. Though there are many standard detecting techniques available, most of the patients die in a year of diagnosis. Hence, it is essential to find an alternative approach for the early recognition of lung cancer, which altogether improves the odds of endurance. This paper addresses the classification and recognition task to detect lung tumor in the early stages of Computerized Tomography (CT) scans for lungs. Lung Image Database Consortium and Infectious Disease Research Institute (LIDC/IDRI) lung image dataset is used for this purpose. Convolutional-Neural-Network (CNN) based deep learning (DL) approach is proposed for the recognition and classification from the CT scan images and the results are compared with the classical machine learning algorithms in the literature with respect to standard evaluation metrics. It is additionally observed that, the Deep CNN classifier has performed better than all the other three traditional classification algorithm in the literature and viewed as the viable classifier.
Convolutional Neural Network approach for the Classification and Recognition of Lung Nodules
05.11.2020
136697 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Lung nodules detection and classification
IEEE | 2005
|Lung Nodules Detection and Classification
British Library Conference Proceedings | 2005
|