2021/02/18 by Antonio Carlos Silva‐Filho, Sara Raquel Dutra-Macêdo, Silva-Filho, Antonio Carlos +5
Engineering · Medicine · #ECG Monitoring and Analysis #FOS: Biological sciences #FOS: Computer and information sciences #Heart Rate Variability and Autonomic Control #Machine Learning (cs.LG) #Non-Invasive Vital Sign Monitoring #Quantitative Methods (q-bio.QM)
paper · pdf · doi:10.48550/arxiv.2102.09410
openalex publication_date 2021/02/18 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Aim: to evaluate the ability of the nonlinear 24-HRV as a predictor of MI\nusing Machine Learning Methods: The sample was composed of 218 patients divided\ninto two groups (Healthy, n=128; MI n=90). The sample dataset is part of the\nTelemetric and Holter Electrocardiogram Warehouse (THEW) database, from the\nUniversity of Rochester Medical Center. We used the most common ML algorithms\nfor accuracy comparison with a setting of 10-fold cross-validation (briefly,\nLinear Regression, Linear Discriminant Analysis, k-Nearest Neighbour, Random\nForest, Supporting Vector Machine, Na "ive Bayes, C 5.0 and Stochastic Gradient\nBoosting). Results: The main findings of this study show that the combination\nof SD1nu + SD2nu has greater predictive power for MI in comparison to other HRV\nindexes. Conclusion: The ML model using nonlinear HRV indexes showed to be more\neffective than the linear domain, evidenced through the application of ML,\nrepresented by a good precision of the Stochastic Gradient Boosting model.\n Keywords: heart rate variability, machine learning, nonlinear domain,\ncardiovascular disease\n