Predição da fração de ejeção do ventrículo esquerdo por meio de algoritmos de deep learning aplicados a sinais de ECG de pacientes chagásicos

Abstract

Objective: To classify the left ventricular ejection fraction of chagasic patients into preserved and non-preserved by using electrocardiography signals. Context: Left ventricular ejection fraction is an important indicator of heart failure and sudden death. To estimate this indicator, echocardiography is necessary, which is usually more expensive and restrictive than electrocardiography. Methods: Initially, we separated the signals into two classes: ejection fraction less than 0.5 (class 1) and ejection fraction greater than or equal to 0.5 (class 2). We used a Tukey’s boxplot to separate noisy beats from non-noisy ones based on their duration. Next, we applied an LSTM (Long-short Term Memory) network to classify sets of 200 beats of each signal. Finally, we applied some methods (mode, sum of probabilities, artificial neural network and LSTM) to obtain a class for the entire signal using the network outputs. Results: We obtained, as the best result, an accuracy of 0.79 and a F1-score of 0.79. Conclusion: We obtained satisfactory results. However, we believe that them can be improved by a more sophisticated beat selection method and a more robust LSTM model.