Suppr超能文献

一种新型计算机心律失常监测系统的临床评估

Clinical evaluation of a new computerized arrhythmia monitoring system.

作者信息

Hultgren H N, Shettigar U R, Specht D F

出版信息

Heart Lung. 1975 Mar-Apr;4(2):241-51.

PMID:1039979
Abstract

A study has been made of the accuracy of a computerized system of arrhythmia monitoring in the coronary-care unit of the Palo Alto Veterans Administration Hospital. The system provides beat-to-beat analysis of cardiac rhythm and QRS morphology and stores information for display of incidence of abnormalities and trend curves. The variables monitored by the system include: (1) heart rate, (2) early beats, (3) late beats, (4) abnormal QRS waveforms, and combinations of 2, 3, and 4. Accuracy analysis was performed using a standard ECG trace with a simultaneous record of computer classification of each QRS complex on a dual-channel recorder. Thirty-four patients with various cardiac arrhythmias were studied and 188,000 QRS complexes were evaluated. The number of true abnormal beats (abnormal in timing or waveform) recorded by the ECG was 12,194 out of 188,000 or 6.5 per cent. Computer classification accuracy was as follows: false positives (computer diagnosis abnormal but ECG diagnosis normal), 0.10 per cent (191 complexes) of all complexes and 1.56 per cent of true positives. False negatives (computer diagnosis normal but ECG diagnosis abnormal), 0.70 per cent (84 complexes) of abnormal beats. Computer monitoring and analysis of seven variables in cardiac rhythm and QRS waveform abnormalities can be performed with a high degree of accuracy using a compact computer system suitable for intensive-care units.

摘要

对帕洛阿尔托退伍军人管理局医院冠心病监护病房心律失常监测计算机系统的准确性进行了一项研究。该系统提供心律和QRS波形态的逐搏分析,并存储信息以显示异常发生率和趋势曲线。该系统监测的变量包括:(1)心率,(2)早搏,(3)晚搏,(4)异常QRS波形,以及2、3和4的组合。使用标准心电图记录进行准确性分析,同时在双通道记录仪上记录每个QRS波群的计算机分类。对34例患有各种心律失常的患者进行了研究,共评估了188000个QRS波群。在188000个记录的QRS波群中,心电图记录的真正异常搏动(时间或波形异常)数量为12194个,占6.5%。计算机分类准确性如下:假阳性(计算机诊断异常但心电图诊断正常),占所有波群的0.10%(191个波群),占真正阳性的1.56%。假阴性(计算机诊断正常但心电图诊断异常),占异常搏动的0.70%(84个波群)。使用适合重症监护病房的紧凑型计算机系统,可以高度准确地对心律和QRS波形异常中的七个变量进行计算机监测和分析。

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验