Classification of Breast Cancer Histopathological Images Using KAZE Features
Breast cancer (BC) is a public health problem of first importance, being the second most common cancer worldwide. BC represents 30.4% of all new cancer cases in the European female population. The diagnosis and differential diagnosis of BC is based on the clinical presentations, physical examinations combined with imaging studies, and confirmed by histopathologic findings. Pathologists’ examination is a time-consuming analysis, susceptible to an interpretation bias mainly caused by the experience of the pathologist and the decrease of attention due to fatigue. Currently, computer-aided detection and diagnosis techniques applied to digital images are assisting the specialists.
In this work, the performance of a pattern recognition system based on KAZE features in combination with Bag-of-Features (BOF) to discriminate between benign and malignant tumours is evaluated on the BreakHis database (7909 images). During the training stage, KAZE keypoints are extracted for every image in the training set. Keypoints are mapped into a histogram vector using K-means clustering. This histogram represents the input to build a binary SVM classifier. In the testing stage, the KAZE keypoints are extracted for every image in the test set, and fed into the cluster model to map them into a histogram vector. This vector is finally fed into the binary SVM training classifier to classify the image.
The experimental evaluation shows the feasibility and effectiveness, in terms of classification accuracy, of the proposed scheme for the binary classification of breast cancer histopathological images with a low magnification factor.
KeywordsBreast cancer Histopathology Image classification Medical imaging KAZE Bag of features
This work was supported by grant TIN2015-67020-P (Spanish “Ministerio de Economía y Competitividad and FEDER funds).
- 1.World Cancer Research Fund International: Cancer facts and figures - Worldwide data. http://www.wcrf.org/int/cancer-facts-figures/worldwide-data. Accessed 18 Jan 2018
- 4.World Health Organization: Breast cancer: prevention and control. http://who.int/cancer/detection/breastcancer/en/index1.html. Accessed 18 Jan 2018
- 13.Zheng, Y., Jiang, Z., Shi, J., Ma, Y.: Retrieval of pathology image for breast cancer using PLSA model based on texture and pathological features. In: 2014 IEEE International Conference on Image Processing (ICIP), pp. 2304–2308. IEEE (2014)Google Scholar
- 14.Spanhol, F.A., Oliveira, L.S., Petitjean, C., Heutte, L.: Breast cancer histopathological image classification using convolutional neural networks. In: International Joint Conference on Neural Networks (IJCNN 2016), pp. 2560–2567 (2016)Google Scholar
- 15.Cruz-Roa, A., Basavanhally, A., González, F., Gilmore, H., Feldman, M., Ganesan, S., Shih, N., Tomaszewski, J., Madabhushi, A.: Automatic detection of invasive ductal carcinoma in whole slide images with convolutional neural networks. Presented at the 20 March 2014Google Scholar
- 18.Rublee, E., Rabaud, V., Konolige, K., Bradski, G.: ORB: an efficient alternative to SIFT or SURF. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 2564–2571 (2011)Google Scholar
- 20.Leutenegger, S., Chli, M., Siegwart, R.Y.: BRISK: binary robust invariant scalable keypoints. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 2548–2555 (2011)Google Scholar
- 21.Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In: Proceedings - 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2005, pp. 886–893 (2005)Google Scholar
- 23.Alahi, A., Ortiz, R., Vandergheynst, P.: FREAK: fast retina keypoint. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 510–517 (2012)Google Scholar
- 25.Demchev, D., Volkov, V., Kazakov, E., Sandven, S.: Feature tracking for sea ice drift retrieval from SAR images. In: 2017 IEEE International Geoscience and Remote Sensing Symposium (IGARSS), pp. 330–333. IEEE (2017)Google Scholar
- 26.Inoue, R., Goto, T., Hirano, S.: Authenticity inspection by image recognition using feature point matching. In: 2017 IEEE 6th Global Conference on Consumer Electronics (GCCE), pp. 1–2. IEEE (2017)Google Scholar
- 27.Utaminingrum, F., Kumiawan, T.A., Fauzi, M.A., Wihandika, R.C., Adikara, P.P.: Adaptive human tracking for smart wheelchair. In: 5th International Symposium on Computational and Business Intelligence, ISCBI 2017, pp. 10–13 (2017)Google Scholar
- 28.Chen, Y.S., Chien, J.C., Lee, J.D.: KAZE-BOF-based large vehicles detection at night. In: 2016 International Conference on Communication Problem-Solving, ICCP 2016, pp. 2–3 (2016)Google Scholar
- 29.Spanhol, F.A., Cavalin, P.R., Oliveira, L.S., Petitjean, C., Heutte, L.: Deep features for breast cancer histopathological image classification. In: IEEE International Conference on Systems, Man, and Cybernetics, pp. 1868–1873 (2017)Google Scholar