Detection of lung injury using 4D-CT chest images

This paper proposes a novel framework for the identification of the radiation-induced lung injury (RILI) after radiation therapy (RT) using 4D computed tomography (CT). The proposed methodology involves elastic image registration; segmentation of the lung fields; extraction of appearance and functional features; and classification of the lung tissues. The first step locally aligns the consecutive phases of the respiratory cycle using an elastic image registration approach based on descent minimization of the sum of squared difference similarity metric. Secondly, lung fields are segmented using a hybrid framework that integrates an adaptive shape prior model, a first-order intensity model, and a second order homogeneity descriptor of the lung tissues. Next, regional features that describe different types of lung functionality (e.g., ventilation and elasticity) are estimated from the segmented lungs. Finally, a random forest classifier has been applied to distinguish between injured and normal lung tissues. The proposed framework has been tested on data sets collected from 13 patients who had undergone RT treatment. Experimental results demonstrate the promise of the proposed framework for the identification of the injured lung region, and thus hold the promise as a valuable tool for early detection of RILI.