Automatic Liver Lesion Extraction from CT Images Based on Distance Regularized Level Set Evolution and Region Growing

Abstract:

Current diagnosis with computed tomography (CT) imaging relies heavily on doctors’ clinical experience and it is difficult to accurately identify and localize lesions from thousands of CT images. Therefore, computer-aided diagnosis with automatic lesion extraction will be helpful for doctors in the diagnosis of liver diseases. In this paper, we proposed a new method for automatic liver lesion extraction from CT images by combining DRLSE (distance regularized level set evolution) and region growing. The method was applied in abdominal CT images with a single liver cancerous lesion and multiple hemangioma lesions at different locations. The results demonstrated the feasibility of our method for automatic lesion extraction with improved diagnostic accuracy and time efficiency.

• Periodical: Applied Mechanics and Materials (Volumes 556-562)
• Pages: 4924-4928
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.556-562.4924
• Online since: May 2014
• Authors: Feng Lian Gao, Lian Fen Huang, Jia Kun Wang, Hai Tao Shuai, Jian Jun Sun*, Yue Huang
• Keywords: Automatic Lesion Extraction, Distance Regularized Level Set Evolution, Liver Segmentation, Region Growing

References

[1] S.S. Kumar, R.S. Moni and J. Rajeesh: SIViP Vol. 7 (2013), pp.163-172.

[2] O. Zayane, B. Jouini and M.A. Mahjoub: Canadian J. Image Processing & Computer Vision Vol. 2 (2011), pp.92-95.

[3] S. Osherand and J.A. Sethian: J. Comput. Phys Vol. 79 (1988), pp.12-49.

[4] V. Caselles, T. Catteand and T. Coll: Numer. Math. Vol. 66 (1993), pp.1-31.

[5] R. Malladi, J.A. Sethian and B.C. Vemuri: IEEE Trans. Pattern Anal. Machine Intell. Vol. 17 (1995), pp.158-175.

DOI: 10.1109/34.368173

[6] V. Caselles, R. Kimmel and G. Sapiro: Int J. Comput. Vision Vol. 22 (1997), pp.61-79.

[7] C. Li, C.Y. Xu and M.D. Fox: IEEE Trans. Image Processing Vol. 19 (2010), pp.3243-3254.