An approach to classification of oral cancer using Softmax Discriminant Classifier
- 1 October 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 420-423
Abstract
One of the most commonly occurring malignant neoplasm worldwide is oral cancer. Various risk factors are associated with the development of oral cancer. For the Oral Squamous Cell Carcinoma (OSCC) to develop, excessive smoking and high amount of alcohol consumption are the major contributing factors. Men tend to have a higher amount of risk for oral cancer than women. Even after the oral cancer is successfully treated, sometimes a relapse of the disease can occur easily and therefore early identification and classification of risk level of oral cancer is quite significant. In this work, Softmax Discriminant Classifier (SDC) is utilized to classify the oral cancer and the performance metrics are evaluated in terms of Performance Index, Classification Accuracy, Sensitivity and Specificity. Results show that an average classification accuracy of 97.29% along with an average Performance Index of 93.93% is obtained.Keywords
This publication has 7 references indexed in Scilit:
- Performance Comparison of Oral Cancer Classification with Gaussian Mixture Measures and Multi Layer Perceptron2017
- Performance analysis of linear layer neural networks for oral cancer classification2017
- Softmax Discriminant Classifier2011
- Developing Prognostic Systems of Cancer Patients by Ensemble ClusteringBioMed Research International, 2009
- Advances in oral cancer detection using optical coherence tomographyIEEE Journal of Selected Topics in Quantum Electronics, 2005
- A novel wavelet neural network based pathological stage detection technique for an oral precancerous conditionJournal of Clinical Pathology, 2005
- IMPROVING DETECTION OF PRECANCEROUS AND CANCEROUS ORAL LESIONSThe Journal of the American Dental Association, 1999