Automatic ECG signal quality assessment

Abstract

The quality assessment of signal has been a research topic for many years, as it is mainly related to the problem of the false alarms. Automatic quality detection/assessment and classification of signals can play a vital role in the development of robust unsupervised electrocardiogram (ECG). The development of efficient algorithms for the quality control of ECG recordings is essential to improve healthcare now. ECG signal can be intermixed with many kinds of unwanted noises. It is an important task to assess the quality of the ECG signal for further biomedical inspections. To make that happen, we made an algorithm that is efficient and uses some basic quality features to classify the ECG signals. It is a very effective way to acquire a good quality ECG signal in real-time by unskilled personnel for instance in rural areas there is not enough expertise in this field. By using this method, they can quickly know if the ECG signal is acceptable or unacceptable for further inspections. The method is used to assess the quality of the ECG signals in the training set of the Physionet/Computing in Cardiology Challenge 2011, giving a correct interpretation of the quality of the ECG signals of 93.08% which corresponded to a sensitivity of 96.53% and a specificity of 86.76%.