A Brief History of Computer-Assisted Electrocardiography

Abstract

This paper reviews the history of the development of computer methods for ECG interpretation. Selected highlights are presented which indicate how technological advances have paralleled the growth of the application of the technique to a point where globally over 100 million ECGs per annum are now interpreted by computer.

Computer Decision Support as a Source of Interpretation Error: The Case of Electrocardiograms

“…3 Previous work supports the notion that the accuracy of the EKG expert system approaches the accuracy of cardiologists, 4 and most researchers believe that EKG expert systems are therefore helpful to the physician. 3,5,6 Laks and Selvester 7 state that in their experience, ''physicians aided by computers produce the best interpretation of EKGs.'' However, few studies have quantified this effect, and none have addressed the effect that incorrect CI has on physician interpretive accuracy.…”

“…A common opinion regarding the CI of EKGs is that even if the interpretation is not correct, it still provides useful information. Macfarlane 5 suggests that the CI ''at the very least provides a second opinion which can be accepted or rejected by a physician.'' Whether this second opinion is always a benefit to the physician is unproven; although two studies have examined how accurate physicians are in accepting correct computer generated advice, 3,6 no studies have examined how physicians use incorrect computer advice.…”

Epidemiological classification of acute myocardial infarction: time for a change?

“…The workload of ECG classification can be minimized by automating the ECG coding in epidemiological studies [25][26][27] . However, there is no consensus of optimal interpretation algorithms for that purpose [28] .…”

Quality Assurance in Cardiac Disease Diagnostic using Computerised Feature Extraction of ECG Signal

“…There were no common definitions of waves, no standards for measurement or diagnostic classification, and no uniform terminology for reporting, transmission and processing of data. This has created a situation whereby large difference result in measurements by different computer programs and hampers the exchange of diagnostic criteria and interpretation results [ 4]. In addition, more and more microcomputer-based interpretative ECG machines are being put on the market without any prior independent validation.…”