Research On Breast Cancer Histopathological Images Recognition Based On Feature Fusion And Nuclei Segmentation

Breast cancer is one of the highest cancer deaths in the world.Pathological examination is the gold standard for breast cancer diagnosis.The recognition of histopathological images of breast cancer has attracted a lot of attention in the field of medical image processing.The Bioimaging 2015 dataset was used to recognize breast cancer histopathological images,and the dataset was classified into two types: cancer and non-cancer in this dissertation.And two automatic recognition models of breast cancer histopathological images based on machine learning methods are proposed.The experiments show that the proposed models have good recognition performance and can obtain high recognition accuracy,the details are as follows:（1）A model of breast cancer histopathological images recognition based on feature fusion is proposed.In the experiments,the features of gray-level co-occurrence matrix in four directions,the wavelet features and Tamura texture features were extracted from the stainedseparated breast cancer histopathological images.And the color features of the original images were extracted according to the color auto-correlogram.At the same time,the gray-level cooccurrence matrix features in the horizontal direction before stain separation were extracted as a supplement to the texture information.Finally,the extracted features were fused and input into the support vector machine classifier to realize the cancer recognition.,and the recognition accuracy is 83.33% on the Bioimaging 2015 dataset.（2）A model of breast cancer histopathological images recognition based on two-stage nuclei segmentation strategy is proposed.The two-stage nuclei segmentation strategy,that is,a method of watershed segmentation based on histopathological images after stain separation,is proposed to make the dataset recognized to be the carcinoma and non-carcinoma recognition.Firstly,stain separation is performed on breast cancer histopathological images.Then the marker-based watershed segmentation method is used for images obtained from stain separation to achieve the nuclei segmentation target.Next,the completed local binary pattern is used to extract texture features from the nuclei regions（images after nuclei segmentation）,and color features were extracted by using the color auto-correlation method on the stain-separated images.Finally,the two kinds of features were fused and the support vector machine was used for carcinoma and non-carcinoma recognition.The experimental results show that the two-stage nuclei segmentation strategy proposed in this paper has significant advantages in the recognition of carcinoma and non-carcinoma,and the recognition accuracy arrives at 91.67%on the Bioimaging 2015 dataset.The proposed method is also applied to the ICIAR 2018 dataset to realize the automatic recognition of carcinoma and non-carcinoma,and the recognition accuracy arrives at 92.50%.