Automatic Breast Cancer Grading of Histological Images Based on Colour and Texture Descriptors
The early diagnosis of breast cancer is extremely important to save lives, but breast cancer diagnosis and prediction is very complex and time consuming. In this article we propose a CAD tool for automated malignancy assessment of breast tissue histological images into four classes: normal, benign, in situ and invasive. The problem is very complex, since histological images exhibit a highly variable appearance, even within the same malignancy level. We compute a features vector related to nuclei, colour regions and textures for each image that serves as an input to a Support Vector Machine (SVM) classifier with a quadratic kernel. System performance has been measure as its classification accuracy using 10-fold cross-validation within an initial set of 400 images. Our approach yields good results with an overall accuracy of 79.2%, and outperforms several other state-of the art algorithms.
KeywordsBreast cancer Computer-aided diagnosis Digital pathology Pattern recognition and classification Tissue malignancy
This work was supported by the Government of Spain [grant number TEC2014-53103-P and TEC2017-82807-P].
- 1.World Cancer Research Fund International. http://www.wcrf.org/int/cancer-facts-figures/data-specific-cancers/breast-cancer-statistics
- 2.Sarmiento, A., Fondón, I.: Breast Cancer Diagnosis CAD (2018). http://personal.us.es/sarmiento/downloads/
- 3.Li, X., Plataniotis, K.N.: Color model comparative analysis for breast cancer diagnosis using H and E stained images. In: SPIE 9420, Medical Imaging 2015: Digital Pathology, vol. 9420 (2015). https://doi.org/10.1117/12.2079935
- 4.Chekkoury, A., Khurd, P., Ni, J., Bahlmann, C., Kamen, A., Patel, A., Grady, L., Singh, M., Groher, M., Navab, N., Krupinski, E., Johnson, J., Graham, A., Weinstein, R.: Automated malignancy detection in breast histopathological images. In: SPIE 8315, Medical Imaging 2012: Computer-Aided Diagnosis, vol. 8315, pp. 831515-1–831515-13 (2012). https://doi.org/10.1117/12.911643
- 6.Spanhol, F.A., Oliveira, L.S., Petitjean, C., Heutte, L.: Breast cancer histopathological image classification using convolutional neural networks. In: Proceedings of International Joint Conference on Neural Networks, pp. 2560–2567 (2016). https://doi.org/10.1109/IJCNN.2016.7727519
- 7.Wang, D., Khosla, A., Gargeya, R., Irshad, H., Becktitle, A.H.: Deep learning for identifying metastatic breast cancer. arXiv:1606.05718 (2016)
- 8.Weil, B., Han, Z., He, X., Yin, Y.: Deep learning model based breast cancer histopathological image classification. In: Proceedings of 2nd IEEE International Conference on Cloud Computing and Big Data Analysis, pp. 348–353 (2017). https://doi.org/10.1109/ICCCBDA.2017.7951937
- 10.Ohashi, T., Al Aghbari, Z., Makinouchi, Z.: Fast segmentation of texture image regions based on hill-climbing. In: Proceedings of IEEE Pacific Rim Conference on Communications, Computers and Signal Processing, (PACRIM), vol. 2, pp. 848–851 (2003)Google Scholar
- 11.Costa, A.F., Humpire-Mamani, G., Traina, A.J.M.: An efficient algorithm for fractal analysis of textures. In: 25th SIBGRAPI Conference on Graphics, Patterns and Images, Ouro Preto, pp. 39–46 (2012). https://doi.org/10.1109/SIBGRAPI.2012.15
- 14.Di Giovanni, P., Ahearn, T.S., Semple, S.I.K., Lovell, L.M., Miller, I., Gilbert, F.J., Redpath, T.W., Heys, S.D., Staff, R.T.: The biological correlates of macroscopic breast tumour structure measured using fractal analysis in patients undergoing neoadjuvant chemotherapy. Breast Cancer Res. Treat. 133(3), 1199–1206 (2001). https://doi.org/10.1007/s10549-012-2014-8
- 15.ICIAR 2018 Grand Challenge on Breast Cancer Histology (BACH). https://iciar2018-challenge.grand-challenge.org/home/