A real-time quality monitoring system for optimal recording of 12-lead resting ECG. (April 2017)

• Record Type: Journal Article
• Authors: Jekova, Irena; Krasteva, Vessela; Leber, Remo; Schmid, Ramun; Twerenbold, Raphael; Reichlin, Tobias; Müller, Christian; Abächerli, Roger
• Is Part Of: Biomedical signal processing and control. Volume 34(2017)
• Page Start: 126
• Page End: 133
• Publication Date: 2017-04
• Subjects: ECG signal quality index -- Signal-to-noise ratio -- Optimal snapshot time -- Adaptive quality threshold; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2017.01.009
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: Quality monitoring system detects the earliest start of diagnostically reliable ECG. Details about global snapshot quality measures of 12-lead ECG are revealed. Adaptive quality threshold is optimized and validated on ECGs from emergency unit. At least 85% of ECGs would be triggered at potentially the best quality. At least 55% of ECGs would be triggered not late than 2.5 s after their best quality. Abstract: Minimizing the impact of artifacts prior to the start of the ECG recording is an approach for providing a diagnostically reliable data. A solution to this problem is a continuous feedback of the ECG quality and prompt start of the recording at potentially the best quality. The real-time lead quality monitoring library (LQMLib) is introduced to trigger the recording of 10 s resting 12-lead ECG at the optimal snapshot moment (the earliest in time, the best in quality). The triggering condition considers the Snapshot quality (signal-to-noise ratio of the most noisy 4 s segment within 10 s) exceeding an adaptive quality threshold (AQT). The optimal AQT (descending from 85% down to 60% over 1 min) is validated on two independent clinical datasets from an emergency department, including 267/385 standard 12-lead ECGs. The test-validation LQMLib performance is: (84.7–87.2)% of ECGs would be triggered at their maximal Snapshot quality; (31.2–33.1)% at the optimal snapshot time (±2.5 s); (25.7–29.3)% would be started earlier, typical for high quality ECGs with progressively increasing supra-threshold Snapshot quality; (37.2–43.1)% would be recorded with a delay >2.5 s, typical for low quality ECGs with sub-threshold maximal Snapshot quality that is not validated as potentially the best quality until AQT criterion declines later within 1 min.