Automated Delineation of Costophrenic Recesses on Chest Radiographs
The lung image segmentation using a model-based approach is a challenge owing to the sheer complexity and variability of the lung shape in a given data set. As a part of our effort to segment the lungs, we report a method to delineate the costophrenic (CP) recess without the human intervention. Active shape model (ASM) is used to point to the probable area of the CP recess, and a prior knowledge-based processing delineates the CP recess and hence determines the angle. The proposed method is fast and shows satisfactory results. It is intended to be used as a preprocessing step in segmenting the lungs’ contour. The proposed method also can be used to initialize the model contour in any other ASM-based lung segmentation algorithms. The algorithm was tested on 45 non-nodule lung images from the JSRT database. An average accuracy of 87.02% is achieved. A comparison of the results of proposed method and gold standard which is obtained by manual delineation is given.
KeywordsCostophrenic angle Active shape model Lung segmentation Computer-aided diagnosis Sensitivity Specificity Jaccard index
- 1.Cootes, T., Taylor, C., Cooper, D., & Graham, J. (1995). Active shape models—their training and application. Computer Vision and Image Understanding, 61, 38–59.Google Scholar
- 3.Paragios, N. & Deriche, R. (1998). Geodesic active regions for texture segmentation. INRIA, Sophia Antipolis, France, Res. Rep. 3440.Google Scholar
- 4.Shi, Y., Q, F., Xue, Z., Chen, L., Ito, K., Matsuo, H., & Shen, D. (2008). Segmenting lung fields in serial chest radiographs using both population-based and patient-specific shape statistics. IEEE Transactions on Medical Imaging, 27(4), 481–494./ https://doi.org/10.1109/tmi.2007.908130.CrossRefGoogle Scholar
- 6.Wan Ahmad, W. S. H. M., & Ahmad Fauzi, M. F., & Zaki, W. (2015). Abnormality detection for infection and fluid cases in chest radiograph (pp. 62–67). https://doi.org/10.1109/elecsym.2015.7380815.
- 7.Campadelli, P., & Casiraghi, E. (2005). Lung field segmentation in digital postero-anterior chest radiographs. In S. Singh, M. Singh, C. Apte, & P. Perner (Eds.), Pattern recognition and image analysis (vol. 3687, pp. 736–745). Lecture Notes in Computer Science. Springer, Heidelberg, Germany.Google Scholar
- 8.Abi-Nahed, J., Jolly, M. P., & Yang, G. Z. (2006). Robust active shape models: a robust, generic and simple automatic segmentation tool. In R. Larsen, M. Nielsen, J. Sporring (Eds.), Medical Image Computing and Computer-Assisted Intervention—MICCAI.Google Scholar
- 9.De la Torre, F., & Black, M. J. (2003). A framework for robust subspace learning. International Journal of Computer Vision, 54(1–3), 117–142.Google Scholar
- 11.Shiraishi, J., Katsuragawa, S., Ikezoe, J., Matsumoto, T., Kobayashi T., Komatsu, K., et al. (2000). Development of a digital image database for chest radiographs with and without a lung nodule: Receiver operating characteristic analysis of radiologists’ detection of pulmonary nodules. AJR 174, 71–74.Google Scholar
- 12.van Ginneken, B., Stegmann, M. B. & Loog, M. (2006). Segmentation of anatomical structures in chest radiographs using supervised methods: A comparative study on a public database. Medical Image Analysis, 10(1), 19–40.Google Scholar