Ranking and classification of monotonic emphysema patterns with a multi-class hierarchical approach

Emphysema has distinct and well-defined visually apparent CT patterns called centrilobular and panlobular emphysema. Existing studies concentrated on the classification of these patterns but they have not looked at the complete evolution of this disease as the destruction of lung parenchyma progresses from normal lung tissue to mild, moderate, and severe disease with complete effacement of the lung architecture. In this paper, we discretize this continuous process into five classes of increasing disease severity and construct a training set of 1161 CT patches. We exploit three solutions to this monotonic multi-class classification problem: a global rankSVM for ranking, hierarchical SVM for classification and a combination of these two, which we call a hierarchical rankSVM. Results showed that both hierarchical approaches were computationally efficient. The classification accuracies were slightly better for hierarchical SVM. However, in addition to classification, ranking approaches also provided a ranking of patterns, which can be utilized as a continuous disease progression score. In terms of the classification accuracy and ratio of pair-wise constraints satisfied, hierarchical rankSVM outperformed the global rankSVM.