Study validates reduction of ECG analysis time when using Cardiologs to detect major arrhythmias vs conventional solution, with similar accuracy
The latest study using the Cardiologs’ Holter Platform entitled “Evaluation of an Ambulatory ECG Analysis Platform Using Deep Neural Networks in Routine Clinical Practice”, was based on a total of 1,000 24-hour ECG recordings collected from 3 tertiary hospitals.
The Cardiologs Holter Platform¹, a clinically-validated arrhythmia diagnostic software, cloud-based, vendor-neutral and powered by artificial intelligence was compared to SyneScope²’s conventional analysis platform in their ability to identify 5 major rhythm abnormalities: pause, ventricular tachycardia, atrial fibrillation/flutter/tachycardia, high-grade atrioventricular block, and high burden of premature ventricular complex (>10%).
Results showed that, when using Cardiologs to analyze the recordings, the analysis time was 26.6% shorter when compared with SyneScope.
Furthermore, there were no statistically significant differences between both platforms when it came to sensitivity and specificity to detect predefined abnormalities except for AFib and ventricular tachycardia (VT). The DNN platform significantly improved sensitivity with no significant change in specificity for atrial fibrillation (98% [96–100] versus 91% [85–96]; P=0.01) and ventricular tachycardia (97% [89– 100] versus 68% [55–79]; P<0.001).
This is the first-to-date performance evaluation of a DNN-based platform for Holter analysis using a large data set of 1,000 24-hour ECG Holters led by experts in a hospital setting. Holters were representative of recordings done in clinical routine and therefore no selection was done. These findings demonstrate that the use of a deep learning approach can help classify a broad range of distinct arrhythmias from ambulatory ECGs with high diagnostic performance.
Dr. Fiorina, Electrophysiologist at the Institut Cardiovasculaire Paris Sud (ICPS) and lead author on the study said: “Since we started using Cardiologs at the ICPS, time saved translated directly into more than two days of electrophysiologists time saved per month, a time that we were able to reinvest to diagnose more patients.”
Today, healthcare professionals are facing numerous challenges with a growing number of patients with heart diseases requiring ECG diagnostics due to growing prevalence of risk factors and aging population. Additionally, we expect ECG data volumes to dramatically increase in the coming years from the fast adoption of ECG analysis solutions embedded in smartwatches and handhelds.
Artificial Intelligence holds a promise to provide accurate and time-efficient Holter analysis, so healthcare professionals can focus on the meaningful episodes where their expertise is needed.
Read the entire publication here.
References
1. The Cardiologs Holter Platform is a medical device intended for use by qualified healthcare professionals for the assessment of arrhythmias using ECG data in subjects over 18 years of age. Class IIa in Europe (CE2797) in compliance with the Medical Device Directive (MM 93/42/EEC amended by 2007/47/EC). Class II in the USA according to the 510K clearance.
2. SyneScope® is a registered trademark by Microport.