Heartbeat Classification by Random Forest With a Novel Context Feature: A Segment Label

Abstract

Objective: Physicians use electrocardiograms (ECG) to diagnose cardiac abnormalities. Sometimes they need to take a deeper look at abnormal heartbeats to diagnose the patients more precisely. The objective of this research is to design a more accurate heartbeat classification algorithm to assist physicians in identifying specific types of the heartbeat.

Methods and procedures: In this paper, we propose a novel feature called a segment label, to improve the performance of a heartbeat classifier. This feature, provided by a Convolutional Neural Network, encodes the information surrounding the particular heartbeat. The random forest classifier is trained based on this new feature and other traditional features to classify the heartbeats.

Results: We validate our method on the MIT-BIH Arrhythmia dataset following the inter-patient evaluation paradigm. The proposed method is competitive with other similar works. It achieves an accuracy of 0.96, and F1-scores for normal beats, ventricular ectopic beats, and Supra-Ventricular Ectopic Beats (SVEB) of 0.98, 0.93, and 0.74, respectively. The precision and sensitivity for SVEB are 0.76 and 0.78, which outperforms the state-of-the-art methods.

Conclusion: This study demonstrates that the segment label can contribute to precisely classifying heartbeats, especially those that require rhythm information as context information (e.g. SVEB). Clinical impact: Using a medical devices embedding our algorithm could ease the physicians' processes of diagnosing cardiovascular diseases, especially for SVEB, in clinical implementation.

Keywords: Convolutional neural network; ECG classification; heartbeat classification; machine learning; mutual information random forest.

FIGURE 1.

Illustration of the overall methodology.

FIGURE 2.

Difference between heartbeat classification and ECG classification. Left side is a single heartbeat with heartbeat label NOR(normal hearbeat). Right side is a period of ECG with ECG label NSR(normal sinus rhythm).

FIGURE 3.

Illustration of QRS width, QRS width in half level, and QRS width in quarter level of one heartbeat. The definition is originally from , here half level and quarter level is calculated between the highest signal strength and lowest signal strength of QRS complex.

FIGURE 4.

Illustration of ECG segmentation without overlap (left) and with half overlap (right). The upper one are the original ECG signals, and the rests are ECG segments got from the records.

FIGURE 5.

Model architecture of ecgclf_c, which is a ECG recording classifier in Fig. 1 serving as a segment label extractor.

FIGURE 6.

Hyperparameter tuning for number of decision trees based on leave-one-patient-out cross-validation on macro-averaged F1-scores of two proposed models, and , which are introduced in Section IV-C.