Classification of Pathological Images Using Convolution Neural Networks

  • Rajashekhargouda C. Patil
Keywords: Oral Cancer, Pathological images, Convolutional Neural Network

Abstract

Cancer can now be counted in the deceases with high mortality rate. Oral cancer is the cancer originating or affecting the oral cavity. Oral cavity consists of the inner part of the open mouth which is visible. Gold standard available for the detection of the oral cancer is through analysis of the microscopic obtained from the Hemotoxilin and Eosin (H&E) staining of the tissue biopsy images. In this paper trying to automate the classification of the above mentioned images with the help of Convolution Neural Networks and Support Vector Machines. The collection used for training the CNN for the feature extraction consists of 300 cancerous images and 60 normal images. Through the proposed approach we achieved the Sensitivity of 80 percent and the specificity of 60 percent.

References

[1] R. Sankaranarayanan, K. Ramadas, G. Thomas et al., “Effect of screening on oral cancer mortality in Kerala, India: a clusterrandomised controlled trial,” The Lancet, vol. 365, no. 9475, pp. 1927–1933, 2005.
[2] Santhosh Kumar Caliaperoumal, R. Vezhavendhan, PriyaVendhan and Uma Devi. “Comparison of Modified Papanicolaou and Hematoxylin and Eosin Stain in Demonstration of Keratin Pearl and Individual Cell Keratin in Oral Squamous Cell Carcinoma”. Int.J.Curr.Microbiol.App.Sci. 2016; 5(7): 558-564.
[3] Chbicheb S, Akerzoul N, Wady WE (2015) Oral Squamous Cell Carcinoma in Moroccan Population: A Cartographic Study. Cosmetol & Oro Facial Surg 1: 102. doi:10.4172/jcofs.1000102.
[4] Yosinski, Jason, et al. "How transferable are features in deep neural networks?." Advances in neural information processing systems. 2014.
[5] Garcia-Gasulla, Dario, et al. "An Out-of-the-box Full-network Embedding for Convolutional Neural Networks." arXiv preprint arXiv:1705.07706 (2017).
[6] Parikh, Rajul et al. “Understanding and Using Sensitivity, Specificity and Predictive Values.” Indian Journal of Ophthalmology 56.1 (2008): 45–50. Print.
Published
2018-04-15
How to Cite
Patil, R. C. (2018, April 15). Classification of Pathological Images Using Convolution Neural Networks. ASIAN JOURNAL FOR CONVERGENCE IN TECHNOLOGY (AJCT ) -UGC LISTED, 4(I). Retrieved from http://asianssr.org/index.php/ajct/article/view/461
Section
Electronics and Telecommunication