Title: Patient-specific early classification of multivariate observations
Authors: Mohamed F. Ghalwash; Dušan Ramljak; Zoran Obradović
Addresses: Center for Data Analytics and Biomedical Informatics, Temple University, Philadelphia, PA, USA; Mathematics Department, Faculty of Science, Ain Shams University, Cairo, Egypt ' Center for Data Analytics and Biomedical Informatics, Temple University, Philadelphia, PA, USA ' Center for Data Analytics and Biomedical Informatics, Temple University, Philadelphia, PA, USA
Abstract: Early classification of time series has been receiving a lot of attention recently. In this paper we present a model, which we call the Early Classification Model (ECM), that allows for early, accurate and patient-specific classification of multivariate observations. ECM is comprised of an integration of the widely used Hidden Markov Model (HMM) and Support Vector Machine (SVM) models. It attained very promising results on the datasets we tested it on: in one set of experiments based on a published dataset of response to drug therapy in Multiple Sclerosis patients, ECM used only an average of 40% of a time series and was able to outperform some of the baseline models, which needed the full time series for classification. In the set of experiments tested on a sepsis therapy dataset, ECM was able to surpass the standard threshold-based method and the state-of-the-art method for early classification of multivariate time series.
Keywords: bioinformatics; patient-specific classification; early classification; multivariate time series; hybrid models; hidden Markov model; HMM; support vector machine; SVM; gene expression; sepsis therapy; early diagnosis.
DOI: 10.1504/IJDMB.2015.067955
International Journal of Data Mining and Bioinformatics, 2015 Vol.11 No.4, pp.392 - 411
Received: 27 Feb 2013
Accepted: 25 Mar 2013
Published online: 12 Mar 2015 *