Brandt, PMoodley, DPillay, AWSeebregts, CJDe Oliveira, T2014-10-092014-10-092014Brandt, P, Moodley, D, Pillay, A.W, Seebregts, C.J and De Oliveira, T. 2014. An investigation of classification algorithms for predicting HIV drug resistance without genotype resistance testing. In: Foundations of Health Information and Engineering Systems (FHIES), Macau, China, 21-23 August 2013. Post-proceedings of the 2013 Symposium on the Foundations of Health Information and Engineering Systems (FHIES), as published as part of Springer's LNCS series, were published in 2014.http://link.springer.com/chapter/10.1007%2F978-3-642-53956-5_16http://hdl.handle.net/10204/7717Foundations of Health Information and Engineering Systems (FHIES), Macau, China, 21-23 August 2013. Post-proceedings of the 2013 Symposium on the Foundations of Health Information and Engineering Systems (FHIES), as published as part of Springer's LNCS series, were published in 2014. Abstract only added.The development of drug resistance is a major factor impeding the efficacy of antiretroviral treatment of South Africa’s HIV infected population. While genotype resistance testing is the standard method to determine resistance, access to these tests is limited in low-resource settings. In this paper we investigate machine learning techniques for drug resistance prediction from routine treatment and laboratory data to help clinicians select patients for confirmatory genotype testing. The techniques, including binary relevance, HOMER, MLkNN, predictive clustering trees (PCT), RAkEL and ensemble of classifier chains were tested on a dataset of 252 medical records of patients enrolled in an HIV treatment failure clinic in rural KwaZulu-Natal in South Africa. The PCT method performed best with a discriminant power of 1.56 for two drugs, above 1.0 for three others and a mean true positive rate of 0.68. These methods show potential for application where access to genotyping is limited.enHIVTreatment failureMachine learningMulti-label classificationClinical decision supportArticleBrandt, P., Moodley, D., Pillay, A., Seebregts, C., & De Oliveira, T. (2014). An investigation of classification algorithms for predicting HIV drug resistance without genotype resistance testing. http://hdl.handle.net/10204/7717Brandt, P, D Moodley, AW Pillay, CJ Seebregts, and T De Oliveira "An investigation of classification algorithms for predicting HIV drug resistance without genotype resistance testing." (2014) http://hdl.handle.net/10204/7717Brandt P, Moodley D, Pillay A, Seebregts C, De Oliveira T. An investigation of classification algorithms for predicting HIV drug resistance without genotype resistance testing. 2014; http://hdl.handle.net/10204/7717.TY - Article AU - Brandt, P AU - Moodley, D AU - Pillay, AW AU - Seebregts, CJ AU - De Oliveira, T AB - The development of drug resistance is a major factor impeding the efficacy of antiretroviral treatment of South Africa’s HIV infected population. While genotype resistance testing is the standard method to determine resistance, access to these tests is limited in low-resource settings. In this paper we investigate machine learning techniques for drug resistance prediction from routine treatment and laboratory data to help clinicians select patients for confirmatory genotype testing. The techniques, including binary relevance, HOMER, MLkNN, predictive clustering trees (PCT), RAkEL and ensemble of classifier chains were tested on a dataset of 252 medical records of patients enrolled in an HIV treatment failure clinic in rural KwaZulu-Natal in South Africa. The PCT method performed best with a discriminant power of 1.56 for two drugs, above 1.0 for three others and a mean true positive rate of 0.68. These methods show potential for application where access to genotyping is limited. DA - 2014 DB - ResearchSpace DP - CSIR KW - HIV KW - Treatment failure KW - Machine learning KW - Multi-label classification KW - Clinical decision support LK - https://researchspace.csir.co.za PY - 2014 T1 - An investigation of classification algorithms for predicting HIV drug resistance without genotype resistance testing TI - An investigation of classification algorithms for predicting HIV drug resistance without genotype resistance testing UR - http://hdl.handle.net/10204/7717 ER -