Learning to Detect 3D Rectal Tubes in CT Colonography Using a Global Shape Model

Abstract

The rectal tube (RT) is a common source of false positives (FPs) in computer-aided detection (CAD) systems for CT colonography. In this paper, we present a novel and robust bottom-up approach to detect the RT. Probabilistic models, trained using kernel density estimation (KDE) on simple low-level features, are employed to rank and select the most likely RT tube candidate on each axial slice. Then, a shape model, robustly estimated using Random Sample Consensus (RANSAC), infers the global RT path from the selected local detections. Our method is validated using a diverse database, including data from five hospitals. The experiments demonstrate a high detection rate of the RT path, and when tested in a CAD system, reduce 20.3% of the FPs with no loss of CAD sensitivity.