A novel multi-threshold method for nodule detection in lung CT

Multi-slice computed tomography (MSCT) is a valuable tool for lung cancer detection thanks to its ability to identify non-calcified nodules of small size (from about 3 mm). Due to the large number of images generated by MSCT, there is much interest in developing computer-aided detection (CAD) systems that could assist radiologists in the lung nodule detection task. A complete multistage CAD system, including lung boundary segmentation, regions of interest (ROIs) selection, feature extraction and false positive (FP) reduction is presented. The selection of ROIs is based on a multi-threshold surface-triangulation approach. Surface triangulation is performed at different threshold values, varying from a minimum to a maximum value in a wide range. The system performance was tested on an independent set of 23 low-dose MSCT scans coming from the Pisa Italung-CT center and on 83 scans made available by the Lung Image Database Consortium (LIDC) annotated by four expert radiologists. On the Italung-CT test set, for nodules having a diameter greater than or equal to 3 mm, the system achieved a 84% and 71% sensitivity at a false positive/scan rate of 10 and 4 respectively. For nodules having a diameter greater 45 than or equal to 4 mm, the sensitivities were 97% and 80% at a false positive/scan rate of 10 and 4 respectively. On the LIDC data set, the system achieved a 79% sensitivity at a false positive/scan rate of 4 in the detection of nodules with a diameter greater than or equal to 3 mm that have been annotated by all four radiologists. 50 Introduction Lung cancer is the leading cause of cancer deaths. The 5-year survival rate is estimated to be only 16% [1]. The survival rate increases up to 49% for cases detected when the disease is still localized; however, only 16% of lung cancers are diagnosed at this early stage [1]. Screening tests for lung cancer focus on trying to detect the disease at an earlier and more curable stage, in particular for high-risk 55 individuals. In the past, chest radiography and sputum cytology were investigated as modalities for lung cancer screening [2,3], showing a limited effectiveness in reducing lung cancer mortality. Lowdose helical computed tomography (CT) has provided promising results in the detection of early-stage lung cancer [4-6]. The large number of images that need to be interpreted by the radiologists in CT screening stimulated the development of various CAD systems for lung nodules [7-22]. Some of the 60 ongoing CT lung screening trials employ multi-slice CT (MSCT) [23,24] which, compared to singleslice helical CT, provides a higher resolution in the axial direction. The availability of tools for a truly three-dimensional data visualization and analysis is particularly important in order to take full advantage of the MSCT isotropic resolution. A complete CAD system for lung nodule detection, structured in four modules, is presented.