Analysis of ECG Data Using AutoML Frameworks to Predict the Classification of Some Cardiovascular Disease Features

The paper presents an approach to analysing and predicting the classification of biomarkers of cardiovascular patients' diseases based on the processing of electrocardiography (ECG) time series data using AutoML frameworks. The authors emphasise the importance of applying artificial intelligence to solve cardiovascular disease diagnosis problems, and investigate the effectiveness of AutoML for automating the creation of machine learning models. In the course of the work a comparative analysis of frameworks is carried out, their advantages and disadvantages are revealed. The results demonstrate the promise of AutoML in the task of predicting classification from numerical and textual cardiac data, but highlight the need for further work on the interpretability and accuracy of models. The paper describes in detail the steps involved in working with the selected dataset, including data preprocessing using exploratory data analysis and vectorisation, pipeline construction and performance evaluations of the selected AutoML frameworks. A conclusion is made about the use of AutoML frameworks for further work with multimodal data in the field of cardiovascular disease.