Hierarchical Lung Field Segmentation With Joint Shape and Appearance Sparse Learning

Abstract

Lung field segmentation in the posterior-anterior (PA) chest radiograph is important for pulmonary disease diagnosis and hemodialysis treatment. Due to high shape variation and boundary ambiguity, accurate lung field segmentation from chest radiograph is still a challenging task. To tackle these challenges, we propose a joint shape and appearance sparse learning method for robust and accurate lung field segmentation. The main contributions of this paper are: 1) a robust shape initialization method is designed to achieve an initial shape that is close to the lung boundary under segmentation; 2) a set of local sparse shape composition models are built based on local lung shape segments to overcome the high shape variations; 3) a set of local appearance models are similarly adopted by using sparse representation to capture the appearance characteristics in local lung boundary segments, thus effectively dealing with the lung boundary ambiguity; 4) a hierarchical deformable segmentation framework is proposed to integrate the scale-dependent shape and appearance information together for robust and accurate segmentation. Our method is evaluated on 247 PA chest radiographs in a public dataset. The experimental results show that the proposed local shape and appearance models outperform the conventional shape and appearance models. Compared with most of the state-of-the-art lung field segmentation methods under comparison, our method also shows a higher accuracy, which is comparable to the inter-observer annotation variation.

LINKS: Learning-based multi-source IntegratioN frameworK for Segmentation of infant brain images

“…However, the classification is performed for each voxel independently, and also as noticed in our previous work (Wang et al, 2014a), the probability maps obtained by the random forest might introduce artificial anatomical errors in the final segmentation results. To deal with this possible limitation, we impose an anatomical constraint into the segmentation by using sparse representation, which has been employed in many applications (Gao et al, 2012; Shao et al, 2014; Wang et al, 2013b; Wang et al, 2014b). Specifically, by applying the trained classification forests, each training subject i can obtain its corresponding forest-based tissue probability maps $({bold-italic\stackrel{false‒}{I}}_{\mathrm{bold-italici}}=\{I_{WM}^i,I_{GM}^i,I_{CSF}^i\})$.…”

Prediction of standard‐dose brain PET image by using MRI and low‐dose brain [18F]FDG PET images

“…Many studies 37,38 have shown that learning multiple local models would help improve the prediction performance, compared to a single global model. Therefore, in this paper, we learn one regression forest for each brain tissue, i.e., WM, GM, and CSF.…”

Severity quantification of pediatric viral respiratory illnesses in chest X-ray images

“…Previous attempts in the literature for the segmentation of lung field from CXR struggle to accommodate large anatomical and pathological variations found in pediatric CXRs. In addition, state-of-the-art existing methods, such as [8], [9], do not delineate parts of lung field behind aortic arch and apex of heart in CXR and therefore annotate the lung field only partially.…”