Classifying Juxta-pleural pulmonary nodules

Abstract

Lung cancer is a disease of abnormal cells multiplying and growing into a tumor in the human lung. It is the most dangerous and widespread cancer in the world. According to the stage of discovery of cancer cells in the lung, the process of early detection plays a very important and essential role to avoid serious advanced stages to reduce its percentage of distribution. Our lung cancer detection system basically detects and recognizes Juxta-pleural pulmonary nodules; which are attached to the wall of the lung. It is done in 4 stages such as obtaining ROI (Region Of Interest), Segmentation, Feature extraction and Classification. CT (Computed Tomography) is considered to be the best modality for the diagnosis of Lung cancer. ROI can be selected either manually or automatically. Automated ROI retrieval is preferred as manual selection is considered to be tedious and time consuming as the operator has to go through the dataset slice by slice and frame by frame. Ray-casting algorithm is used to segment nodule and neural networks are used to classify the nodules appropriately.