OBJECTIVE: To classify Right Bundle Branch Block (RBBB),Left Bundle Branch Block (LBBB) and normal ECG signals automatically. METHODS: The MIT-BIH database was used as experimental data sources.The training set and test set were extracted for training and testing network models.Based on convolutional neural network,this paper proposed the core algorithm:sparse connection residual network.Compared the sparse connected residual network with classic network models,then evaluated the recognition effect of the model. RESULTS: The accuracy of the test set the MIT-BIH database was 95.2%,the result is better than classic network models. CONCLUSIONS The algorithm proposed in this paper can assist doctors in the diagnosis of heart block related disease and place a high value on clinical application.
Algorithms ; Arrhythmias, Cardiac ; diagnostic imaging ; Bundle-Branch Block ; diagnostic imaging ; Electrocardiography ; Humans ; Neural Networks (Computer)

BACKGROUNDUsing tissue Doppler imaging and conventional echocardiographic technique, we examined the cardiac function and synchronicity in individuals with isolated right bundle branch block (RBBB) or left bundle branch block (LBBB) and assessed the relationship between QRS duration and synchronicity.

METHODSSubjects with isolated RBBB (n = 20), LBBB (n = 10) and normal controls (n = 20) were studied with conventional echocardiography and tissue Doppler imaging. The difference between aortic and pulmonary preejection intervals was defined as interventricular delay. Parameters in septum and lateral wall were measured using tissue Doppler imaging, including peak sustained systolic velocity (S(M)), peak early (E(M)) and late (A(M)) diastolic velocities as well as time to peak velocities (T(S), T(E) and T(A)).

RESULTSSubjects with LBBB had lower S(M) and longer T(S) than did the RBBB and control groups (P < 0.05, P < 0.001 respectively). A significant difference was observed in E(M), being the lowest in the LBBB and the highest in the control group (P < 0.05). Moreover, T(E) was longer in the LBBB group compared with the other two groups (P < 0.001). Both A(M) and T(A) were similar among three groups (P > 0.05). In the bundle branch block groups, one ventricle lagged about 40 ms behind the other. A significant correlation was found between interventricular delay and QRS duration (r = 0.713, P < 0.001).

CONCLUSIONSCardiac ventricles were not well synchronized with one ventricle lagging about 40 ms behind the other in subjects with LBBB or RBBB, even though only LBBB group showed barely perceptible, impaired cardiac function. In addition, QRS duration and cardiac asynchronicity were positively correlated.

Adult ; Aged ; Bundle-Branch Block ; diagnostic imaging ; physiopathology ; Diastole ; Echocardiography, Doppler ; methods ; Electrocardiography ; Female ; Heart ; physiopathology ; Humans ; Male ; Middle Aged ; Systole

Left anterior fascicular block (LAFB) is a heart disease identifiable from an abnormal electrocardiogram (ECG). It has been reported that LAFB is associated with an increased risk of heart failure. Non-specific intraventricular conduction delay due to the lesions of the conduction bundles and slow cell to cell conduction has also been considered as another cause of heart failure. Since the location and mechanism of conduction delay have notable variability between individual patients, we hypothesized that the impaired conduction in the ventricular myocardium may lead to abnormal ECGs similar to LAFB ECG patterns. To test this hypothesis, based on a computer model with a three dimensional whole-heart anatomical structure, we simulated the cardiac exciting sequence map and 12-lead ECG caused by the block in the left anterior fascicle and by the slowed conduction velocity in the ventricular myocardium. The simulation results showed that the typical LAFB ECG patterns can also be observed from cases with slowed conduction velocity in the ventricular myocardium. The main differences were the duration of QRS and wave amplitude. In conclusion, our simulations provide a promising starting point to further investigate the underlying mechanism of heart failure with LAFB, which would provide a potential reference for LAFB diagnosis.
Adult ; Bundle-Branch Block/diagnostic imaging* ; Computer Simulation ; Electrocardiography ; Heart/diagnostic imaging* ; Heart Atria/diagnostic imaging* ; Heart Conduction System/physiopathology* ; Heart Failure/diagnostic imaging* ; Heart Ventricles/diagnostic imaging* ; Humans ; Male ; Models, Anatomic ; Models, Theoretical ; Muscle Cells ; Myocardium ; Phantoms, Imaging ; Poisson Distribution