A Novel Hepatocellular Carcinoma Image Classification Method Based on Voting Ranking Random Forests.

Bingbing Xia,Huiling Liu,Huiyan Jiang,Dehui Yi

doi:10.1155/2016/2628463

Abstract

This paper proposed a novel voting ranking random forests (VRRF) method for solving hepatocellular carcinoma (HCC) image classification problem. Firstly, in preprocessing stage, this paper used bilateral filtering for hematoxylin-eosin (HE) pathological images. Next, this paper segmented the bilateral filtering processed image and got three different kinds of images, which include single binary cell image, single minimum exterior rectangle cell image, and single cell image with a size of n⁎n. After that, this paper defined atypia features which include auxiliary circularity, amendment circularity, and cell symmetry. Besides, this paper extracted some shape features, fractal dimension features, and several gray features like Local Binary Patterns (LBP) feature, Gray Level Cooccurrence Matrix (GLCM) feature, and Tamura features. Finally, this paper proposed a HCC image classification model based on random forests and further optimized the model by voting ranking method. The experiment results showed that the proposed features combined with VRRF method have a good performance in HCC image classification problem.

Highlights

As we all know, liver cancer is a common disease and, just like some other diseases, the global morbidity and mortality rates have significant increases recently
When it comes to conventional random forests, the proposed voting ranking random forests (VRRF) method has a certain improvement
This paper proposed a classification method for three kinds of segmented single cell image, which include single binary cell image (SBCI), single minimum exterior rectangle cell image (SMERCI), and single cell image with a size of n ∗ n (SNSCI)

Summary

Introduction

Liver cancer is a common disease and, just like some other diseases, the global morbidity and mortality rates have significant increases recently This problem is a huge challenge that mankind will face for a long time [1, 2]. Downs et al [4] proposed a decision support tool based on fuzzy ARTMAP for the diagnosis of breast cancer. This method solved the subjective inaccuracy cognizance problem with objective fuzzy category membership by using fuzzy recognition theory and realized HCC image diagnosis. This paper proposed a novel random forests classification model based on voting ranking method; besides, this paper defined three innovative atypia features including auxiliary circularity, amendment circularity, and cell symmetry. The proposed VRRF method showed good performance in hepatocellular carcinoma image classification and strong robustness

Hepatocellular Carcinoma Image Classification Method

Results evaluation

Results and Discussion

Conclusion