3D pulmonary nodules detection using fast marching segmentation

Pulmonary nodule detection is an important step in lung cancer detection because nodules arethe alert signal of lung cancer. The early detection of them can hence increase the patient’ssurvival rates. This paper proposes an automated computer aided diagnosis system fordetection of pulmonary nodules based on three dimensional (3D) structures. Lungparenchyma segmentation using fast marching method was employed. A simple thresholdingtechnique is used to extract candidate nodules from the segmented lung parenchyma. A 3Dimage of nodule candidates is reconstructed by mean of stacked 2D images. To find theconnected voxels of a blob, a 3D connected component labelling is used. Features extractedfrom each blob are then fed into the classifier. The random forest algorithm has been invokedfor nodule and non-nodule classification. The proposed detection methodology can give theaccuracy of 92%.Keywords: lung cancer; pulmonary nodule; fast marching; 3D features; random forestclassifier