Abstract
For both the acquisition of mobile electrocardiogram (ECG) devices and early warning and diagnosis of clinical work, high-quality ECG signals is particularly important. We describe an effective system which could be deployed as a stand-alone signal quality assessment algorithm for vetting the quality of ECG signals. The proposed ECG quality assessment method is based on the simple heuristic fusion and fuzzy comprehensive evaluation of the SQIs. This method includes two modules, i.e., the quantification and extraction of Signal Quality Indexes (SQIs) for different features, intelligent assessment and classification. First, simple heuristic fusion is executed to extract SQIs and determine the following SQIs: R peak detection match qSQI, QRS wave power spectrum distribution pSQI, kurtosis kSQI, and baseline relative power basSQI. Then, combined with Cauchy distribution, rectangular distribution and trapezoidal distribution, the membership function of SQIs was quantified, and the fuzzy vector was established. The bounded operator was selected for fuzzy synthesis, and the weighted membership function was used to perform the assessment and classification. The performance of the proposed method was tested on the database from Physionet ECG database, with an accuracy (Acc) of 97.67%, sensitivity (Se) of 96.33% and specificity (Sp) of 98.33% on the training set. Testing against the test datasets resulted in scores of 94.67, 90.33, and 93.00%, respectively. There's no gold standard exists for determining the quality of ECGs. However, the proposed algorithm discriminates between high- and poor-quality ECGs, which could aid in ECG acquisition for mobile ECG devices, early clinical diagnosis and early warning.
Highlights
With the wide application of mobile ECG in the fields of financial safety, security monitoring, medical insurance, and data confidentiality, ECG recording devices are not limited to professional training staff.From the perspective of mobile ECG collection, most of the available ECG recording devices lack real-time feedback about the signal quality
For any ECG signal to be evaluated, the Signal Quality Indexes (SQIs) quality evaluation mechanism based on fuzzy comprehensive evaluation can be used to obtain the ECG quality assessment results according to the above formula (33)
The results of the comparison of Fuzzy comprehensive evaluation with the simple heuristic fusion of the SQIs with Database D test are shown in Tables 4, 5 below: For simple heuristic fusion of the SQIs, when the number of SQIs increases from 4 to 5, the accuracy of the database D1 and D2 is not well optimized, even in its sensitivity (Se) and specificity (Sp)
Summary
With the wide application of mobile ECG in the fields of financial safety, security monitoring, medical insurance, and data confidentiality, ECG recording devices are not limited to professional training staff. From the perspective of mobile ECG collection, most of the available ECG recording devices lack real-time feedback about the signal quality. It is difficult for non-professionals to collect high-quality ECG signals. For the diagnosis of cardiovascular disease (World Health Organization, 2016), the high costs of primary health care make followup treatment unaffordable To circumvent this problem, many countries transmit real-time ECG data recorded by patients to clinical experts to diagnose patients. We discuss the correlation between ECG signal quality and noise and ECG waveform characteristics to obtain accurate assessment results using simple rules and complex classification techniques. If ECG signals are collected for multiple leads that are independent of each other, each channel can be processed separately
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
