Cardiac disease classification using heart rate signals

V. Mahesh; A. Kandaswamy; C. Vimal; B. S. Sathish

doi:10.1504/ijeh.2010.034173

January 1, 2010International Journal of Electronic Healthcare

Cardiac disease classification using heart rate signals

View at Publisher

Ask AI

Bookmark

Share

Why the study?

Does combining linear and nonlinear measures of heart rate variability improve the classification accuracy of cardiac diseases compared to either measure alone?

Key Points

To assess whether combining linear and nonlinear heart rate variability parameters enhances the classification accuracy of cardiac diseases using machine learning algorithms.
Extracted linear (time domain and frequency domain) and nonlinear heart rate variability features from standard electrocardiogram recordings in PhysioNet databases.
Trained and evaluated Random Forests, Logistic Model Tree, and Multilayer Perceptron Neural Network models using linear, nonlinear, and combined feature sets.
Models trained on combined linear and nonlinear measures achieved higher classification accuracy for cardiac abnormalities than models utilizing either measure set independently.
Classification performance across all three algorithms aligned with standard predictive benchmarks previously established in cardiovascular literature.

Structured PICO

Does combining linear and nonlinear measures of heart rate variability improve the classification accuracy of cardiac diseases compared to either measure alone?

P

Population

ECG data from standard ECG databases in the Physionet website

I

Intervention

Machine learning classification (Random Forests, Logistic Model Tree, Multilayer Perceptron Neural Network) using combined linear and nonlinear measures of heart rate variability

C

Comparator

Classification using linear or nonlinear measures alone

O

Outcome

Accurate classification of cardiac diseases

Combining linear and nonlinear measures of heart rate variability improves the accuracy of machine learning classifiers in detecting cardiac diseases from ECG signals.

Ask about this paper

Answers are grounded in this paper and cite their sources.

Implication

May enhance ECG-based ML detection of cardiac disease; leaves open prospective validation before clinical use.

Abstract

Heart rate and Heart Rate Variability (HRV) are important measures that reflect the state of the cardiovascular system. HRV analysis has gained prominence in the field of cardiology for detecting cardiac abnormalities. This paper presents the study made on the use of linear (time domain and frequency domain) and nonlinear measures of heart rate variability for accurate classification of certain cardiac diseases. Three different classifiers, viz. Random Forests, Logistic Model Tree and Multilayer Perceptron Neural Network have been used for the classification. Data for use in this work has been obtained from the standard ECG databases in the Physionet website. Classification has been attempted using linear parameters, nonlinear parameters and combined. The classification results indicate that the combination of linear and nonlinear measures is a better indicator of heart diseases than linear or nonlinear measures alone. The results obtained by this study are comparable with those obtained with other techniques cited in the literature.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Mahesh et al. (2010) studied this question.

synapsesocial.com/papers/6a7139f678a11c550e0b47e3 https://doi.org/10.1504/ijeh.2010.034173

View at Publisher

Ask AI

Bookmark

Share

Cardiac disease classification using heart rate signals

Key Points

Structured PICO

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Cardiac disease classification using heart rate signals

Key Points

Structured PICO

Ask about this paper

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading