This work is devoted to the study of the possibility of determining heart diseases on the basis of 13 categorical and numerical signs. We present a detailed analysis of the dataset, including dividing the data into training and test samples, dividing features into numerical and categorical, applying 4 different classification algorithms, checking the quality of the model using two techniques – delayed sampling and cross-validation. To assess the quality of the model, we pay attention to the value of the recall metric and the error matrix built on the test dataset from the deferred sample or on each test fold when using cross-validation. The results of the study are important both for a deep understanding of the relationship between certain medical indicators and heart disease, and for the development of effective methods for predicting them in the presence of individual symptoms.

Keywords: cardiovascular diseases, classification task, quality metrics, cross-validation, recall, machine learning, random forest

Automatic classification of electrocardiogram signals will allow providing timely medical care to patients when providing first aid. Neural network models of electrocardiogram signal classification, including the stage of preliminary signal processing, allow increasing the accuracy of classifying electrocardiograms into a particular category of arrhythmia. The paper presents a computational method for preliminary processing of electrocardiogram signals, including noise reduction using discrete wavelet transform and extraction of morphological features using frequency analysis methods. The results of modeling the classification of 12-channel electrocardiogram signals using the stage of their preliminary processing showed an increase in classification accuracy by 23.2% compared to classification without preliminary signal processing.

Keywords: electrocardiogram signal classification, long-term short-term memory neural network, metadata, signal preprocessing wavelet transform, spectral analysis, PhysioNet Computing in Cardiology Challenge 2021

Diseases of the cardiovascular system are the main cause of death in the world. The main way to diagnose diseases of the cardiovascular system is to take an electrocardiogram of the patient. Automatic processing of electrocardiogram signals will allow doctors to quickly identify heart problems in a patient. This article presents a method for calculating the detail vector for neural network processing of a twelve-channel electrocardiogram signal. Adding a detail vector to the electrocardiogram signals improves the classification accuracy to 87.50%. The proposed method can be used to automatically classify two or more channel electrocardiogram signals.

Keywords: electrocardiogram, recurrent neural network, neural network with long-term short memory, detailing vector, PhysioNet Computing in Cardiology Challenge 2021