Indian Journal of Science and Technology

Open Access Open Access  Restricted Access Subscription Access

Treatment Response Classification in Randomized Clinical Trials: A Decision Tree Approach


Affiliations
1 Department of Statistics, National Institute for Research in Tuberculosis, (ICMR), Chennai-600 031, India
 

Decision Trees are a subfield of machine learning technique within the larger field of artificial intelligence. It is a supervised learning technique for classification and prediction. The decision trees are widely used for outcome prediction under various treatments for disease cure, prevention, toxicity and relapse. The aim of the paper is to compare the decision tree algorithms in classifying tuberculosis patient's response under randomized clinical trial condition. Classification of patient's responses to treatment is based on bacteriological and radiological methods. Three decision tree approaches namely C4.5, Classification and regression trees (CART), and Iterative dichotomizer 3 (ID3) methods were used for the classification of response. The result shows that C4.5 decision tree algorithm performs better than CART and ID3 methods.

Keywords

Decision Tree, Data Mining, Cart. C4.5, ID3
User

  • Anyanwu M N,Shiva SG (2009) Comparative analysis of serial decision treeclassification algorithms. International Journal of Computer Science and Security 3(3): 230-240

  • Breiman L, Friedman J, Olshen R, Stone C (1984) Classification and RegressionTrees. Wadsworth International Group, Belmont, CA

  • Das R (2010) A comparison of multiple classification methods for diagnosis ofParkinson disease. Expert Systems with Applications 37: 1568–1572

  • Gerald LB, Tang S, Bruce F, Redden D, Kimerling ME, Brook N, Dunlap N, BaileyWC (2002) A Decision Tree for Tuberculosis Contact Investigation. American Journal of Respiratory and Critical Care Medicine166: 1122-1127

  • Han J, Kamber M (2006) Data Mining: Concepts and techniques. Morgan Kaufmann,2nd edition

  • James KE, White RF, Kraemer HC (2005) Repeated split sample validation to assesslogistic regression and recursive partitioning: an application to the predictionof cognitive impairment. Statistics in Medicine 24(19): 3019-3035

  • Jawahar MS (2004) Current trends in chemotherapy of tuberculosis. Indian Journal ofMedical Research 120: 398- 417

  • Kim YS (2010) Performance evaluation for classification methods: comparativesimulation study. Expert Systems with Applications 37(3): 2292-2306

  • Kohavi R (1995) A study of cross-validation and bootstrap for accuracy estimationand model selection. Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence2 (12): 1137–1143(Morgan Kaufmann, San ateo)

  • Lewis RJ (2000) An introduction to classification and regression tree (CART)analysis. Paper presented at the annual meeting of the Society for Academic Emergency Medicine, San Francisco, CA

  • Li H, Sun, J and Wu, J. (2010) Predicting business failure using classification andregression tree: An empirical comparison with popular classical statistical methods and top classification mining methods. Expert Systems with Applications 37 (8): 5895-5904

  • Mello FCQ, Bastos LGV, Soares SLM, Rezende VM, Conde MB, Chaisson R E,Kritski AL, Netto AR, Werneck GL (2006) Predicting smear negative pulmonary tuberculosis with classification trees and logistic regression: a crosssectional study.BMC Public Health6(43): 1-8

  • Podgorelec V, Kokol P, Stiglic B, Rozman I (2002) Decision trees: An overview andtheir use in medicine.J Med Syst26(5): 445-463

  • Pramanik S, Md. R Islam, Md. J Uddin (2010) Pattern Extraction, Classification andcomparison between attribute selection measures. International Journal of Computer Science and Information Technologies 1(5): 371-375

  • Quinlan JR (1979) Discovering rules by induction from large collections of examples.Expert Systems in the Micro Electronic Age, Edinburgh University Press, 168–201

  • Quinlan JR (1986) Induction of Decision Trees. Machine Learning 1: 81-106

  • Quinlan JR (1993) C4.5: Programs for Machine Learning. Morgan KaufmannPublishers, San Mateo, CA

  • Quinlan JR (1996) Improved use of continuous attributes in C4.5. Journal of ArtificialIntelligence Research 4: 77-90

  • Schlipköter U, Flahault A (2010) Communicable diseases: achievements andchallenges for public health. Public Health Reviews32:90-119

  • Steinberg D, Colla P (1997) CART - Classification and Regression Trees, San Diego,California: Salford Systems, http://www.salfordsystems.com/cart.php.

  • Tuberculosis Research Centre, Madras (1983) Study of chemotherapy regimens of 5and 7 months duration and the role of corticosteroids in the treatment of sputum-positive patients with pulmonary tuberculosis in south India. Tubercle 64: 73-91

  • Ture M, Kurta I, Kurumb AT, Ozdamarc K (2005) Comparing classificationtechniques for predicting essential hypertension. Expert Systems with Applications 29: 583-588

  • Utgoff P, Brodley C (1990) An Incremental Method for Finding Multivariate Splitsfor Decision Trees. Machine Learning: Proceedings of the Seventh International Conference, pp.58

  • Weka reference WEKA: http://www.cs.waikato.ac.nz/~ml/weka/


Abstract Views: 537

PDF Views: 152




  • Treatment Response Classification in Randomized Clinical Trials: A Decision Tree Approach

Abstract Views: 537  |  PDF Views: 152

Authors

P. Venkatesan
Department of Statistics, National Institute for Research in Tuberculosis, (ICMR), Chennai-600 031, India
N. R. Yamuna
Department of Statistics, National Institute for Research in Tuberculosis, (ICMR), Chennai-600 031, India

Abstract


Decision Trees are a subfield of machine learning technique within the larger field of artificial intelligence. It is a supervised learning technique for classification and prediction. The decision trees are widely used for outcome prediction under various treatments for disease cure, prevention, toxicity and relapse. The aim of the paper is to compare the decision tree algorithms in classifying tuberculosis patient's response under randomized clinical trial condition. Classification of patient's responses to treatment is based on bacteriological and radiological methods. Three decision tree approaches namely C4.5, Classification and regression trees (CART), and Iterative dichotomizer 3 (ID3) methods were used for the classification of response. The result shows that C4.5 decision tree algorithm performs better than CART and ID3 methods.

Keywords


Decision Tree, Data Mining, Cart. C4.5, ID3

References





DOI: https://doi.org/10.17485/ijst%2F2013%2Fv6i1%2F30563