2020/10/29 by Giulia Silveri, Silveri, Giulia, Marco Merlo +13
Computer Science · Engineering · Medicine · #Artificial intelligence #Artificial neural network #Blood pressure #Cardiology #Computer science #ECG Monitoring and Analysis #Ejection fraction #Electrocardiography #Heart Rate Variability and Autonomic Control #Heart failure #Heart rate #Heart rate variability #Identification (biology) #Internal medicine #Machine learning #Medicine #Non-Invasive Vital Sign Monitoring #cs.LG #eess.SP
paper · pdf · doi:10.48550/arxiv.2010.15893
published in arXiv (Cornell University) (Cornell University)
arxiv created 2020/10/29 · openalex publication_date 2020/10/29 · arxiv updated 2020/11/02 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
The diagnosis of heart diseases is a difficult task generally addressed by an\nappropriate examination of patients clinical data. Recently, the use of heart\nrate variability (HRV) analysis as well as of some machine learning algorithms,\nhas proved to be a valuable support in the diagnosis process. However, till\nnow, ischemic heart disease (IHD) has been diagnosed on the basis of Artificial\nNeural Networks (ANN) applied only to signs, symptoms and sequential ECG and\ncoronary angiography, an invasive tool, while could be probably identified in a\nnon-invasive way by using parameters extracted from HRV, a signal easily\nobtained from the ECG. In this study, 18 non-invasive features (age, gender,\nleft ventricular ejection fraction and 15 obtained from HRV) of 243 subjects\n(156 normal subjects and 87 IHD patients) were used to train and validate a\nseries of several ANN, different for number of input and hidden nodes. The best\nresult was obtained using 7 input parameters and 7 hidden nodes with an\naccuracy of 98.9% and 82% for the training and validation dataset,\nrespectively.\n