A Novel Method of HCC Pathological Images Recognition based on Morphological Distribution of Liver Cell Cords
- 10.2991/bbe-16.2016.76How to use a DOI?
- Liver Cell Cords, HCC, Pattern Recognition, Pathological Image.
The morbidity of hepatocellular carcinoma (HCC), which has become a significant killer to human life, is getting higher over the years. Not only the shape and structure of liver cells are abnormal in pathological images, but also liver cell cords show some unusual phenomena, such as incrassation and disorder. So our paper proposes a novel method of HCC pathological images recognition based on morphological distribution of liver cell cords, to assist pathologist to diagnose hepatic pathology. In the method, we firstly apply binarization and morphological operation to hepatic pathological images, and screen connected regions with their area and circularity. Adopted regions are considered as candidate centers of liver cell cords (vein center). Secondly we define different features (irregularity feature), according to the number of candidate centers in a pathological image, to recognize how structured liver cell cords are. Finally, we utilize the feature and classifier to identify HCC pathological images. The experiments indicate that the proposed method shows high accuracy of all HCC pathological images in well, moderately, poorly differentiated and normal forms.
- © 2016, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Aoshuang Dong AU - Su Jiang AU - Huiyan Jiang AU - Keming Mao PY - 2016/07 DA - 2016/07 TI - A Novel Method of HCC Pathological Images Recognition based on Morphological Distribution of Liver Cell Cords BT - Proceedings of the 2016 International Conference on Biomedical and Biological Engineering PB - Atlantis Press SP - 482 EP - 493 SN - 2468-5747 UR - https://doi.org/10.2991/bbe-16.2016.76 DO - 10.2991/bbe-16.2016.76 ID - Dong2016/07 ER -