Automated classification of lung bronchovascular anatomy in CT using AdaBoost

Abstract

Lung CAD systems require the ability to classify a variety of pulmonary structures as part of the diagnostic process. The purpose of this work was to develop a methodology for fully automated voxel-by-voxel classification of airways, fissures, nodules, and vessels from chest CT images using a single feature set and classification method. Twenty-nine thin section CT scans were obtained from the Lung Image Database Consortium (LIDC). Multiple radiologists labeled voxels corresponding to the following structures: airways (trachea to 6th generation), major and minor lobar fissures, nodules, and vessels (hilum to peripheral), and normal lung parenchyma. The labeled data was used in conjunction with a supervised machine learning approach (AdaBoost) to train a set of ensemble classifiers. Each ensemble classifier was trained to detect voxels part of a specific structure (either airway, fissure, nodule, vessel, or parenchyma). The feature set consisted of voxel attenuation and a small number of features based on the eigenvalues of the Hessian matrix (used to differentiate structures by shape). When each ensemble classifier was composed of 20 weak classifiers, the AUC values for the airway, fissure, nodule, vessel, and parenchyma classifiers were 0.984 ± 0.011, 0.949 ± 0.009, 0.945 ± 0.018, 0.953 ± 0.016, and 0.931± 0.015 respectively. The strong results suggest that this could be an effective input to higher-level anatomical based segmentation models with the potential to improve CAD performance.

Hierarchical Discriminative Framework for Detecting Tubular Structures in 3D Images

“…These algorithms model the tubular pattern directly and use discriminative classifiers on features describing low-level image characteristics [5,10,20]. The classifier responses are then used in a region growing algorithm to find the tree segmentations.…”

Segmentation of inflamed synovia in multi-modal 3D MRI

“…Other works use more advanced algorithms, such as AdaBoost or Support Vector Machines (SVMs), both widely used by the computer vision community. AdaBoost has been used by [4] to classify voxels in chest CT images, with a features dictionary fixed from the onset, while in [5], the knee cartilage is segmented using an SVM trained on banks of 3D Gabor filters subsequently smoothed by 3D Gaussians. Compared to these approaches, our method has the advantage of a datadependent dictionary which can easily adapt to different tasks, resulting in classifier using relatively few features.…”

Probabilistic branching node detection using hybrid local features

“…These are important problems in medical imaging [8,9,10,11,12,18,19]. Early works usually involve manual or semi-manual efforts, often combined with vessel specific enhancement techniques.…”

“…A similar approach is applied for retinal vessel segmentation in [9]. Machine learning techniques play important roles in some recent systems for vessel anatomy study [11]. In [11], Adaboost [13] is applied on features for classification of lung bronchovascular anatomy.…”

“…Machine learning techniques play important roles in some recent systems for vessel anatomy study [11]. In [11], Adaboost [13] is applied on features for classification of lung bronchovascular anatomy. Tracking-based approaches have also been applied for vessel detection [14,10].…”