Automated Gland Segmentation Leading to Cancer Detection for Colorectal Biopsy Images
Glandular formation and morphology along with the architectural appearance of glands exhibit significant importance in the detection and prognosis of inflammatory bowel disease and colorectal cancer. The extracted glandular information from segmentation of histopathology images facilitate the pathologists to grade the aggressiveness of tumor. Manual segmentation and classification of glands is often time consuming due to large datasets from a single patient. We are presenting an algorithm that can automate the segmentation as well as classification of H and E (hematoxylin and eosin) stained colorectal cancer histopathology images. In comparison to research being conducted on cancers like prostate and breast, the literature for colorectal cancer segmentation is scarce. Inter as well as intra-gland variability and cellular heterogeneity has made this a strenuous problem. The proposed approach includes intensity-based information, morphological operations along with the Deep Convolutional Neural network (CNN) to evaluate the malignancy of tumor. This method is presented to outpace the traditional algorithms. We used transfer learning technique to train AlexNet for classification. The dataset is taken from MCCAI GlaS challenge which contains total 165 images in which 80 are benign and 85 are malignant. Our algorithm is successful in classification of malignancy with an accuracy of 90.40, Sensitivity 89% and Specificity of 91%.
KeywordsColorectal cancer Malignant Benign Glands Segmentation Convolutional neural networks (CNN)
We are thankful to MCCAI GlaS 2015 contest for providing us with the relevant data .
- 3.Xu, J., Madabhushi, A., Janowczyk, A., Chandran, S.: A weighted mean shift, normalized cuts initialized color gradient based geodesic active contour model: applications to histopathology image segmentation. In: Proceedings of SPIE, vol. 7623, April 2016, pp. 76230Y–76230Y–12 (2010)Google Scholar
- 5.Gao, Y., Liu, W., Arjun, S., Zhu, L., Ratner, V., Kurc, T., Saltz, J., Tannenbaum, A.: Multi-scale learning based segmentation of glands in digital colorectal pathology images. In: Proceedings of SPIE, vol. 9791. pp. 97910M–97910M–6 (2016)Google Scholar
- 6.Sirinukunwattana, K., Pluim, J.P.W., Chen, H., Qi, X., Heng, P.-A., Guo, Y.B., Wang, L.Y., Matuszewski, B.J., Bruni, E., Sanchez, U., Böhm, A., Ronneberger, O., Ben Cheikh, B., Racoceanu, D., Kainz, P., Pfeiffer, M., Urschler, M., Snead, D.R.J., Rajpoot, N.M.: Gland segmentation in colon histology images: the GlaS challenge contest, pp. 1–24 (2016)Google Scholar
- 15.Ciresan, D.C., Giusti, A., Gambardella, L.M., Schmidhuber, J.: Mitosis detection in breast cancer histology images with deep neural networks. Med. Image Comput. Comput. Assist. Interv. 16(Pt 2), 411–418 (2013)Google Scholar
- 16.Chen, T., Chefd’hotel, C.: Deep learning based automatic immune cell detection for immunohistochemistry images. In: Wu, G., Zhang, D., Zhou, L., (eds.) Machine Learning in Medical Imaging. (Lecture Notes in Computer Science), vol. 8689, pp. 17–24. Springer International Publishing, Berlin (2014)Google Scholar
- 20.Nasim, A., Hassan, T., Akram, M.U., Hassan, B., et al.: Automated identification of colorectal glands morphology from benign images. In: International Conference on IP, Computer Vision, and Pattern Recognition IPCV’17 (2017)Google Scholar
- 21.Alexnet toolbox at Mathworks [ONLINE] Available at: https://www.mathworks.com/help/nnet/ref/alexnet.html. Accessed 13 April 2018