Detection of strict left bundle branch block by neural network and a method to test detection consistency

Abstract

Objective: To develop an automatic algorithm to detect strict left bundle branch block (LBBB) on electrocardiograms (ECG) and propose a procedure to test the consistency of neural network detections. Approach: The database for the classification of strict LBBB was provided by Telemetric and Holter ECG Warehouse. It contained 10 s ECGs taken from the MADIT-CRT clinical trial. The database was divided into a training dataset (N = 300, strict LBBB = 174, non-strict LBBB = 126) and a test dataset (N = 302, strict LBBB = 156, non-strict LBBB = 146). LBBB-related features were extracted by Philips DXL™ algorithm, selected by a random forest classifier, and fed into a 5-layer neural network (NN) for the classification of strict LBBB on the training dataset. The performance of NN on the test dataset was compared to two random forest classifiers, an algorithm applying strict LBBB criteria, a wavelet-based approach, and a support-vector-machine approach. The consistency of NN’s detection was tested on 549 2 min recordings of the PTB diagnostic ECG database. LBBB annotations are not required to measure consistency. Main results: The performance of NN on the test dataset were sensitivity = 91. 7%, specificity = 85.6% and accuracy = 88.7% (PPV = 87.2%, NPV = 90.6%). The consistency score of strict-LBBB and non-strict-LBBB detection was 0.9341 and 0.9973 respectively. Conclusion: NN achieved the highest specificity, accuracy, and PPV. Using random forest for feature selection and NN for classification increased interpretability and reduced computational cost. The consistency test showed that NN achieved high consistency scores in the detection of strict LBBB. Significance: This work proposed an approach for selecting features and training NN for the detection of strict LBBB as well as a consistency test for black-box algorithms.

Compelling new electrocardiographic markers for automatic diagnosis

“…The results are particularly good for the CLBBB rule, which confirms the potential of omeR in diagnosis. Specifically, the sensitivity and specificity values obtained for the CLBBB classification are similar or even higher than those obtained from approaches proposed by other authors, such as [17][18][19][20][21] . Nevertheless, compar-isons with other rules are not fair as none of them has such universal character or they are not possible to be applied automatically.…”

“…For example, [20] proposed a method based on wavelet analysis, obtaining high and moderate values of sensitivity and specificity (92.9% and 65.1%, respectively). A more accurate diagnosis was achieved in [21] , which combines a random forest classifier and a neural network with sensitivity and specificity values of 91.7% and 88.7%, respectively. Both algorithms were trained and validated using data on 600 patients from the Multicenter Automatic Defibrillator Implantation Trial-Cardiac Resynchronization Therapy (MADIT-CRT) database [22] .…”

Compelling new electrocardiographic markers for automatic diagnosis

“…The results are particularly good for the CLBBB rule, which confirms the potential of omeR in diagnosis. Specifically, the sensitivity and specificity values obtained for the CLBBB classification are similar or even higher than those obtained from approaches proposed by other authors, such as [19,17,18,20,21]. Nevertheless, comparisons with other rules are not fair as none of them has such universal character or they are not possible to be applied automatically.…”

Detection of Bundle Branch Blocks using Machine Learning Techniques

“…They have used various classifiers like kNN, neural networks and SVM. Yang et al [30] presents an automatic algorithm based on 5-layer neural network for the detection of strict left bundle branch block. Our suggested technique has better accuracy than these methods.…”