Detection and station mapping of mediastinal lymph nodes on thoracic computed tomography using spatial prior from multi-atlas label fusion

Lymph nodes play an important role in clinical practice but detection is challenging due to low contrast surrounding structures and variable size and shape. In this paper, we propose a fully automatic method for mediastinal lymph node detection and station mapping on thoracic CT scans. First, lungs are automatically segmented to locate the mediastinum region. Shape features by Hessian analysis, local scale, and circular transformation are computed at each voxel. Spatial prior distribution is determined based on the identification of 11 anatomical structures by using multiatlas label fusion. Shape features and spatial prior are then integrated for lymph node detection. The detected candidates are segmented by curve evolution. Characteristic features are calculated on the segmented lymph nodes and support vector machine is utilized for classification and false positive reduction. We applied our method to 20 patients with 62 enlarged mediastinal lymph nodes. The system achieved a significant improvement with 80% sensitivity at 8 false positives per patient with spatial prior compared to 45% sensitivity at 8 false positives per patient without a spatial prior. With the segmentation of spatial anatomic structures, 88% of mediastinal lymph nodes are correctly mapped to their stations.