PROJECT TITLE :

Lung Nodule Classification With Multilevel Patch-Based Context Analysis - 2014

ABSTRACT:

In this paper, we propose a unique classification technique for the four types of lung nodules, i.e., well-circumscribed, vascularized, juxta-pleural, and pleural-tail, in low dose computed tomography scans. The proposed technique relies on contextual analysis by combining the lung nodule and surrounding anatomical structures, and has three main stages: an adaptive patch-primarily based division is employed to construct concentric multilevel partition; then, a brand new feature set is intended to include intensity, texture, and gradient information for image patch feature description, and then a contextual latent semantic analysis-primarily based classifier is designed to calculate the probabilistic estimations for the relevant images. Our proposed method was evaluated on a publicly on the market dataset and clearly demonstrated promising classification performance.