Feature selection methods for optimizing clinicopathologic input variables in oral cancer prognosis

The incidence of oral cancer is high for those of Indian ethnic origin in Malaysia. Various clinical and pathological data are usually used in oral cancer prognosis. However, due to time, cost and tissue limitations, the number of prognosis variables need to be reduced. In this research, we demonstr...

Full description

Bibliographic Details
Main Authors: Chang, S.W., Kareem, S.A., Kallarakkal, A.F., Merican, A.F., Abraham, M.T., Zain, R.B.
Format: Article
Published: 2011
Subjects:
_version_ 1825718724305354752
author Chang, S.W.
Kareem, S.A.
Kallarakkal, A.F.
Merican, A.F.
Abraham, M.T.
Zain, R.B.
author_facet Chang, S.W.
Kareem, S.A.
Kallarakkal, A.F.
Merican, A.F.
Abraham, M.T.
Zain, R.B.
author_sort Chang, S.W.
collection UM
description The incidence of oral cancer is high for those of Indian ethnic origin in Malaysia. Various clinical and pathological data are usually used in oral cancer prognosis. However, due to time, cost and tissue limitations, the number of prognosis variables need to be reduced. In this research, we demonstrated the use of feature selection methods to select a subset of variables that is highly predictive of oral cancer prognosis. The objective is to reduce the number of input variables, thus to identify the key clinicopathologic (input) variables of oral cancer prognosis based on the data collected in the Malaysian scenario. Two feature selection methods, genetic algorithm (wrapper approach) and Pearson's correlation coefficient (filter approach) were implemented and compared with single-input models and a full-input model. The results showed that the reduced models with feature selection method are able to produce more accurate prognosis results than the full-input model and single-input model, with the Pearson's correlation coefficient achieving the most promising results.
first_indexed 2024-03-06T05:08:44Z
format Article
id um.eprints-2798
institution Universiti Malaya
last_indexed 2024-03-06T05:08:44Z
publishDate 2011
record_format dspace
spelling um.eprints-27982013-11-11T01:27:07Z http://eprints.um.edu.my/2798/ Feature selection methods for optimizing clinicopathologic input variables in oral cancer prognosis Chang, S.W. Kareem, S.A. Kallarakkal, A.F. Merican, A.F. Abraham, M.T. Zain, R.B. R Medicine The incidence of oral cancer is high for those of Indian ethnic origin in Malaysia. Various clinical and pathological data are usually used in oral cancer prognosis. However, due to time, cost and tissue limitations, the number of prognosis variables need to be reduced. In this research, we demonstrated the use of feature selection methods to select a subset of variables that is highly predictive of oral cancer prognosis. The objective is to reduce the number of input variables, thus to identify the key clinicopathologic (input) variables of oral cancer prognosis based on the data collected in the Malaysian scenario. Two feature selection methods, genetic algorithm (wrapper approach) and Pearson's correlation coefficient (filter approach) were implemented and compared with single-input models and a full-input model. The results showed that the reduced models with feature selection method are able to produce more accurate prognosis results than the full-input model and single-input model, with the Pearson's correlation coefficient achieving the most promising results. 2011 Article PeerReviewed Chang, S.W. and Kareem, S.A. and Kallarakkal, A.F. and Merican, A.F. and Abraham, M.T. and Zain, R.B. (2011) Feature selection methods for optimizing clinicopathologic input variables in oral cancer prognosis. Asian Pacific Journal of Cancer Prevention, 12 (10). ISSN 1513-7368, DOI PMID: 22320970. http://www.ncbi.nlm.nih.gov/pubmed/22320970 PMID: 22320970
spellingShingle R Medicine
Chang, S.W.
Kareem, S.A.
Kallarakkal, A.F.
Merican, A.F.
Abraham, M.T.
Zain, R.B.
Feature selection methods for optimizing clinicopathologic input variables in oral cancer prognosis
title Feature selection methods for optimizing clinicopathologic input variables in oral cancer prognosis
title_full Feature selection methods for optimizing clinicopathologic input variables in oral cancer prognosis
title_fullStr Feature selection methods for optimizing clinicopathologic input variables in oral cancer prognosis
title_full_unstemmed Feature selection methods for optimizing clinicopathologic input variables in oral cancer prognosis
title_short Feature selection methods for optimizing clinicopathologic input variables in oral cancer prognosis
title_sort feature selection methods for optimizing clinicopathologic input variables in oral cancer prognosis
topic R Medicine
work_keys_str_mv AT changsw featureselectionmethodsforoptimizingclinicopathologicinputvariablesinoralcancerprognosis
AT kareemsa featureselectionmethodsforoptimizingclinicopathologicinputvariablesinoralcancerprognosis
AT kallarakkalaf featureselectionmethodsforoptimizingclinicopathologicinputvariablesinoralcancerprognosis
AT mericanaf featureselectionmethodsforoptimizingclinicopathologicinputvariablesinoralcancerprognosis
AT abrahammt featureselectionmethodsforoptimizingclinicopathologicinputvariablesinoralcancerprognosis
AT zainrb featureselectionmethodsforoptimizingclinicopathologicinputvariablesinoralcancerprognosis