重症监护病人心电导联信号质量评估

Signal quality assessment of ECG in the intensive care unit

目的:研究基于重症监护病人心电导联的信号质量评估算法。方法:通过分析信号的波形特征、统计特性和相互关系导出反映信号质量高低的信号质量指数(Signal Quality Index,SQI),并基于SQI进行病人的心率估计,应用美国麻省理工学院多参数智能重症监护数据库Ⅱ中437例病人的6 000多小时高质量数据和添加的各类心电干扰的数据进行算法评价。结果:SQI随信噪比的降低而减小;SQI与心电搏动检测灵敏度和正检测率呈高度正相关关系;基于信号质量评估算法,在严重干扰存在时仍能提供精确的心率估计。结论:信号质量评估算法可对心电信号质量给出客观的评价。

Objective： To develop an algorithm of signal quality assessment of ECG in the intensive care unit. Methods： The signal quality index （SQI） was obtained by analyzing the morphological and statistical characteristics of each waveform and their relationships. The heart rate was robustly estimated based on SQI. The algorithm was evaluated using more than 6 000 hours of simultaneously acquired ECG from a 437-patient subset of the Multi-Parameter Intelligent Monitoring for Intensive Care II database and adding real ECG noises. Results： The SQI was decreased along with a decrease of the signal noise ratio. The SQI had high-positive correlation with the QRS detective sensitivity and positive predictivity. This method provided an accurate HR estimation even in the presence of high levels of persistent noises and artifacts. Conclusion： The signal quality assessment algorithm could provide an objective estimation of ECG quality.