Lung Nodule Identification

One of the major causes of death in the world for both women and men is lung cancer. The survival rate for a person affected by lung cancer can be increased, if it is detected in its early stages. The detection of the cancer cells that cause the lung cancer, is one of the main concerns in the field of medical image processing. Nodules are one of the most usual signs of lung cancer. Computed Tomography (CT) image is a suitable technique for detecting small pulmonary nodules. This paper proposes a system to detect lung nodule by using CT image. It consists of five stages, Pre-Processing, Lung Region Extraction, Nodule Segmentation, Feature Extraction, and Classification. The slice number which contains even the smallest nodules are also identified. This can help radiologists and doctors to detect lung cancer in early stages. The CT images for the testing of this project are acquired from the Lung Image Database Consortium (LIDC).