A new database with annotations of P waves in ECGs with various types of arrhythmias

OBJECTIVE

The aim of this study is to create a database for the development, evaluation and objective comparison of algorithms for P wave detection in ECG signals. Brno University of Technology ECG Signal Database with Annotations of P-Wave (BUT PDB) is an ECG signal database with marked peaks of P waves annotated by ECG experts. Currently, there are only a few databases of pathological ECG signals with P-wave annotations, and some are incorrect.

APPROACH

The pathological ECG signals used in this work were selected from three existing databases of ECG signals: MIT-BIH Arrhythmia Database, MIT-BIH Supraventricular Arrhythmia Database and Long Term AF Database. The P-wave positions were manually annotated by two ECG experts in all selected signals.

MAIN RESULTS

The final BUT PDB composed of selected signals consists of 50 two-minute, two-lead pathological ECG signal records with annotated P waves. Each record also contains a description of the diagnosis (pathology) present in the selected part of the record and information about positions and types of QRS complexes.

SIGNIFICANCE

The BUT PDB is created for developing new, more accurate and robust methods for P wave detection. These algorithms will be used in medical practice and will help cardiologists to evaluate ECG records, establish diagnoses and save time.

Robust Heartbeat Classification for Wearable Single-Lead ECG via Extreme Gradient Boosting

Wearable electrocardiogram (ECG) monitoring devices have enabled everyday ECG collection in our daily lives. However, the condition of ECG signal acquisition using wearable devices varies and wearable ECG signals could be interfered with by severe noises, resulting in great challenges of computer-aided automated ECG analysis, especially for single-lead ECG signals without spare channels as references. There remains room for improvement of the beat-level single-lead ECG diagnosis regarding accuracy and efficiency. In this paper, we propose new morphological features of heartbeats for an extreme gradient boosting-based beat-level ECG analysis method to carry out the five-class heartbeat classification according to the Association for the Advancement of Medical Instrumentation standard. The MIT-BIH Arrhythmia Database (MITDB) and a self-collected wearable single-lead ECG dataset are used for performance evaluation in the static and wearable ECG monitoring conditions, respectively. The results show that our method outperforms other state-of-the-art models with an accuracy of 99.14% on the MITDB and maintains robustness with an accuracy of 98.68% in the wearable single-lead ECG analysis.

[Establishment of Application Database for Electrocardiograph in Different Scenes]

It is significant to establish scene ECG database which improves the automatic diagnostic function of electrocardiograph under different application scenarios. We built the ECG database in different scene according to the hospital level (grade 3, grade 2, grade 1) and clinical environment (intensive care and acute wards, outpatient clinics and general wards). Sample size was obtained according to the incidence of various ECG diagnoses. The database covers ECG signal, ECG waveform, ECG characteristic values, ECG diagnostic results by experts and clinical information of patients etc. It not only provides important reference for electrocardiograph manufacturers to evaluate and test the parameters of automatic diagnosis under different clinical scene, but also provides valuable scientific research and teaching resources for medical workers.