Oral cancer analysis using machine learning techniques

Title

Oral cancer analysis using machine learning techniques

Creator

Lavanya L.; Chandra J.

Description

Oral cancer staging is most required task of examining treatment and required medication for the patients. In oral cancer staging is of two types, namely, clinical and pathological. TNM (Tumor, Node, and Metastasis) staging is clinical system of predicting oral cancer stages, on the other hand Histology, p63 and podoplanin expressions are pathological staging system which is obtained after biopsy test. These staging systems are used in machine leaning techniques to analyze the different stages of oral cancer. The main aim of the paper is to classify different stages of oral cancer using machine learning techniques. The experimental work is based on clinical and pathological staging system. The data set used for this research work is based on Oral Leukoplakia. The data transformation is applied to standardize the data and the features were extracted using correlation coefficient. The extracted features were classified using Decision tree and random forest which are compared against other popular classification methods like SVM, KNN, MLP and Logistic Regression. From the experimental work, it is found that the various stage classification of oral cancer can be classified efficiently with help of Random Forest and Decision Tree. So the classification of various oral cancers can be performed with help of random forest and Decision Tree. International Research Publication House.

Source

International Journal of Engineering Research and Technology, Vol-12, No. 5, pp. 596-601.

Date

2019-01-01

Publisher

International Research Publication House

Subject

Decision tree; KNN; Machine learning; Oral cancer; OSCC; Prediction; Random forest; SVM; TNM

Coverage

Lavanya L., Department of Computer Science, CHRIST (Deemed to be University), India; Chandra J., Department of Computer Science, CHRIST (Deemed to be University), India

Rights

Restricted Access

Relation

ISSN: 9743154

Format

Online

Language

English

Type

Article

Identifier

https://www.scopus.com/inward/record.uri?eid=2-s2.0-85069983569&partnerID=40&md5=91a9887b0af5454295b0d85a5105732d