Poincare plot analysis has been recognized to provide valuable information in the prognostic stratification of cardiac patients. The parameters provided by the analysis can be used as input for machine learning algorithms in order to distinguish patients in Health, Hypertension, Post-myocardial infarction, Congestive heart failure, Heart transplanted classes. Knime analytics platform was employed to implement decision tree and random forests algorithms. Some evaluation metrics (accuracy, sensitivity and specificity) were computed to assess the final performance. The best accuracy was 86% while the highest specificity was 98.8%. The analysis proved the feasibility of machine learning in predicting patients with different cardiac issues based on parameters extracted through a Poincare analysis.
Machine Learning applied on Poincaré Analyisis to discriminate different cardiac issues
Cesarelli G.;
2020-01-01
Abstract
Poincare plot analysis has been recognized to provide valuable information in the prognostic stratification of cardiac patients. The parameters provided by the analysis can be used as input for machine learning algorithms in order to distinguish patients in Health, Hypertension, Post-myocardial infarction, Congestive heart failure, Heart transplanted classes. Knime analytics platform was employed to implement decision tree and random forests algorithms. Some evaluation metrics (accuracy, sensitivity and specificity) were computed to assess the final performance. The best accuracy was 86% while the highest specificity was 98.8%. The analysis proved the feasibility of machine learning in predicting patients with different cardiac issues based on parameters extracted through a Poincare analysis.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.