Pulmonary Micronodule Detection from 3D Chest CT Sukmoon Chang 1,2 , Hirosh Emoto 3 , Dimitris N. Metaxas 1 , and Leon Axel 4 1 Center for CBIM, Rutgers University, Piscataway, NJ, USA {sukmoon,dnm}@cs.rutgers.edu 2 Computer Science, Penn State Capital College, Middletown, PA, USA 3 National Institute of Information and Communications Technology, Tokyo, Japan jiang@nict.go.jp 4 Department of Radiology, New York University, New York, NY, USA leon.axel@med.nyu.edu Abstract. Computed Tomography (CT) is one of the most sensitive medical imaging modalities for detecting pulmonary nodules. Its high contrast resolution allows the detection of small nodules and thus lung cancer at a very early stage. In this paper, we propose a method for automating nodule detection from high-resolution chest CT images. Our method focuses on the detection of discrete types of granulomatous nod- ules less than 5mm in size using a series of 3D filters. Pulmonary nodules can be anywhere inside the lung, e.g., on lung walls, near vessels, or they may even be penetrated by vessels. For this reason, we first develop a new cylinder filter to suppress vessels and noise. Although nodules usually have higher intensity values than surrounding regions, many malignant nodules are of low contrast. In order not to ignore low contrast nodules, we develop a spherical filter to further enhance nodule intensity values, which is a novel 3D extension of Variable N-Quoit filter. As with most automatic nodule detection methods, our method generates false positive nodules. To address this, we also develop a filter for false positive elimi- nation. Finally, we present promising results of applying our method to various clinical chest CT datasets with over 90% detection rate. 1 Introduction Early detection of lung cancer is critical to improving chances of survival. The five-year survival rate of lung cancer patients is nearly 50% if lung cancer is found at a localized state (i.e., before it has spread to other organs) and can reach 85% if it is diagnosed in an early stage and surgery is possible [1,2]. Once the cancer has spread to other organs, the survival rates decline dramatically— 20% at regional stage and 2.2% at distant stage. Nevertheless, only 15% of lung cancer cases are found at the localized early stage. For early diagnosis of lung cancer, it is critical to detect nodules less than 5mm in size. Various computational methods have been developed and considerable efforts have been made on automating nodule detection from chest radiographs [16]. C. Barillot, D.R. Haynor, and P. Hellier (Eds.): MICCAI 2004, LNCS 3217, pp. 821–828, 2004. c  Springer-Verlag Berlin Heidelberg 2004