Evaluation of machine learning methods for the long-term prediction of cardiac diseases

Author

Schlemmer, Alexander ; Zwirnmann, Henning ; Zabel, Martin ; Parlitz, Ulrich ; Luther, Samuel

Author_Institution

Max Planck Inst. for Dynamics & Self-Organ., Göttingen, Germany

fYear

2014

fDate

25-28 May 2014

Firstpage

157

Lastpage

158

Abstract

We evaluate several machine learning algorithms in the context of long-term prediction of cardiac diseases. Results from applying K Nearest Neighbors Classifiers (KNN), Support Vector Machines (SVM) and Random Forests (RF) to data from a cardiological long-term study suggests that multivariate methods can significantly improve classification results. SVMs were found to yield the best results in Matthews Correlation Coefficient and are most stable with respect to a varying number of features.

Keywords

correlation methods; diseases; electrocardiography; feature extraction; learning (artificial intelligence); medical signal processing; signal classification; support vector machines; Matthews correlation coefficient; SVM; cardiac diseases; cardiological long-term prediction; k-nearest neighbors classifiers; machine learning algorithms; multivariate methods; random forests; support vector machines; Algorithm design and analysis; Cardiac disease; Correlation; Europe; Prediction algorithms; Support vector machines; Vegetation;

fLanguage

English

Publisher

ieee

Conference_Titel

Cardiovascular Oscillations (ESGCO), 2014 8th Conference of the European Study Group on

Conference_Location

Trento

Type

conf

DOI

10.1109/ESGCO.2014.6847567

Filename

6847567