Willcox ME, wait. LBCT02-01. Published in: Heart Rhythm 2021; July 28-31, 2021 (Mixed Conference).

Willcox ME, wait. LBCT02-01. Published in: Heart Rhythm 2021; July 28-31, 2021 (Mixed Conference).

Researchers reported on Heart Rhythm 2021 that long-term continuous ECG monitoring under human supervision is superior to artificial intelligence-based mobile cardiac telemetry monitors to detect arrhythmias.

The detection and preliminary interpretation of the results from the outpatient heart monitor is performed by the device manufacturer, “so we tried to compare the accuracy of the two real-world monitors we use in the clinic,” Mark E. Willcox, an electrophysiologist at Alaska Heart Medical PhD and Vascular Institute, Anchorage, said in a press conference.

The monitor is a mobile heart telemetry system (BioGuardian MCT/CEM, Preventice Solutions) and a long-term continuous ECG monitor (Carnation dynamic monitor, Bardy Diagnostics).

"The fundamental difference between these monitors comes down to how the full disclosure data is analyzed after it reaches the independent data testing facility, which processes the raw data and converts it into interpretable strips," Willcox said. "Preventice’s MCT/CEM does this by using a computer algorithm that scans the full disclosure. If it detects arrhythmias, then these arrhythmias will be interpreted by humans and included in the final report. There is a complete manual review of the long-term ECG. Disclosure of data, extract relevant and important content. There is an algorithm that can support the person and increase sensitivity. In the final report, they are presented differently."

Fifty patients who ordered a 30-day mobile cardiac telemetry system were required to wear long-term continuous ECG monitors at the same time. Two electrophysiologists interpret and rule all reports and compare all ruled arrhythmias. Wilcox said readers of the two companies are not aware of the trial.

Willcox said the results included data from 46 patients (average age 58; 33% male) who wore two monitors for an average of 10.3 days, of which 43% of mobile cardiac telemetry systems defaulted to cardiac event monitoring. Press conference.

According to the researcher, there were 46 arrhythmia adjudicated during simultaneous wearing, of which 24% were diagnosed by the mobile heart telemetry system, and 50% were diagnosed by the long-term continuous ECG monitor (P = .018).

Willcox said that the number of arrhythmia episodes detected during simultaneous wearing was 19 for mobile cardiac telemetry systems and 61 for long-term continuous ECG monitors (P <.001).

He said that the mobile cardiac telemetry system detected 1 arrhythmia missed by the long-term continuous ECG monitor, and the long-term continuous ECG monitor detected 46 arrhythmia missed by the mobile cardiac telemetry system.

"Not all monitors are created equal," Wilcox said at a press conference. "We all know that computers are not as good as humans by reading electrocardiograms, but we are very dependent on them in the outpatient world. We are not saying that one monitor is clinically better than another, because we have not compared clinical utility, only Diagnostic accuracy."

The arrhythmias detected in the study were atrial fibrillation of at least 10 seconds, atrial flutter of at least 10 seconds, at least 20 atrial tachycardia, at least 3 ventricular tachycardia of 100 bpm, and second-degree atrioventricular conduction Block, complete heart block, sinus bradycardia less than 30 bpm for at least 30 seconds, 2:1 sinus node outlet block, atrioventricular nodal reentrant tachycardia and atrioventricular reentrant tachycardia .

Jagmeet P. Singh, MD, PhD

The study showed us that humans can still outperform machines in identifying rhythm disorders. The authors proved that continuous ECG monitoring read by humans performs much better than algorithm-based mobile heart telemetry. The author reiterated this point. When using algorithm-driven monitoring systems, diagnostic accuracy is often overlooked for ease of use and clear reporting.

I think this should keep us vigilant and carefully check the mobile heart telemetry report. We should always ensure that the patient’s symptoms are clinically associated with the ECG strip, because sometimes the ECG read by the algorithm may be false negative.

The study needs to be repeated in a larger population cohort, with more monitors and a variety of patients, with variable baseline ECG characteristics. Mainly to make it more universal.

Nevertheless, the future will still be the judgment of rhythm disorders derived from artificial intelligence or algorithms. We will get there, but it will be repeated. The competition between humans and machines is healthy, but the future is about cooperation.

Send the latest news and education to your inbox

© 2021 Healio All rights reserved.

Send the latest news and education to your inbox

© 2021 Healio All rights reserved.