Artificial Intelligence–Enabled Assessment of the Heart Rate Corrected QT Interval Using a Mobile Electrocardiogram Device

Abstract

Background: Heart rate-corrected QT interval (QTc) prolongation, whether secondary to drugs, genetics including congenital long QT syndrome (LQTS), and/or systemic diseases including SARS-CoV-2-mediated COVID19, can predispose to ventricular arrhythmias and sudden cardiac death. Currently, QTc assessment and monitoring relies largely on 12-lead electrocardiography. As such, we sought to train and validate an artificial intelligence (AI)-enabled 12-lead electrocardiogram (ECG) algorithm to determine the QTc, and then prospectively test this algorithm on tracings acquired from a mobile ECG (mECG) device in a population enriched for repolarization abnormalities. Methods: Using over 1.6 million 12-lead ECGs from 538,200 patients, a deep neural network (DNN) was derived (n = 250,767 patients for training and n = 107,920 patients for testing) and validated (n = 179,513 patients) to predict the QTc using cardiologist over-read QTc values as the gold standard. The ability of this DNN to detect clinically-relevant QTc prolongation (e.g. QTc ≥ 500 ms) was then tested prospectively on 686 genetic heart disease (GHD) patients (50% with LQTS) with QTc values obtained from both a 12-lead ECG and a prototype mECG device equivalent to the commercially-available AliveCor KardiaMobile 6L. Results: In the validation sample, strong agreement was observed between human over-read and DNN-predicted QTc values (-1.76 ± 23.14 ms). Similarly, within the prospective, GHD-enriched dataset, the difference between DNN-predicted QTc values derived from mECG tracings and those annotated from 12-lead ECGs by a QT expert (-0.45 ± 24.73 ms) and a commercial core ECG laboratory [+10.52 ms ± 25.64 ms] was nominal. When applied to mECG tracings, the DNN's ability to detect a QTc value ≥ 500 ms yielded an area under the curve, sensitivity, and specificity of 0.97, 80.0%, and 94.4%, respectively. Conclusions: Using smartphone-enabled electrodes, an AI-DNN can predict accurately the QTc of a standard 12-lead ECG. QTc estimation from an AI-enabled mECG device may provide a cost-effective means of screening for both acquired and congenital LQTS in a variety of clinical settings where standard 12-lead electrocardiography is not accessible or cost-effective.

Portable single-lead electrocardiogram device is accurate for QTc evaluation in hospitalized patients

“…An artificial intelligence (AI)-enabled 12-lead ECG algorithm to determine the QTc, was evaluated in 686 patients with genetic heart disease (50% with long QT syndrome). A strong agreement was observed between a deep neural network-predicted QTc values derived from manual ECG (mECG) tracings and those annotated from 12-lead ECGs by a QT expert (-0.45±24.73 ms), with values very similar to our findings ( 19 ). These results suggest that QTc measured with wearable devices could be applied for ambulatory surveillance, which is likely to increase over time, even after the pandemic has subsided.…”

Patient-specific, re-engineered cardiomyocyte model confirms the circumstance-dependent arrhythmia risk associated with the African-specific common SCN5A polymorphism p.S1103Y: Implications for the increased sudden deaths observed in black individuals during the COVID-19 pandemic

“…In emergency situations, where treatment with any known QT-prolonging drug needs to begin immediately, in-hospital, post-treatment QT monitoring should be performed. With mobile ECG devices capable of accurately assessing the QTc in nearly any setting, 11 consideration of routine pre-and post-treatment QTc monitoring will soon become practically and financially feasible. Due to p.S1103Y-SCN5A's circumstancedependent risk, this is particularly important for patients of African descent with concomitant QT risk factors or those at-risk for metabolic or respiratory acidosis, including those with moderateto-severe manifestations of COVID-19.…”

Community pharmacist use of mobile ECG to inform drug therapy decision making for patients receiving QTc prolonging medications

“…Also, the electrodes should be cleaned with an alcohol‐based sanitizer between each patient use. Artificial intelligence approaches have been successful in accurately calculating QTc interval with the KardiaMobile 6 L device 30 . Future availability of this technology may eliminate the need for manual QTc interval calculations with this mobile ECG device and decrease the overall time required for the pharmacist.…”

“…L device 30. Future availability of this technology may eliminate the need for manual QTc interval calculations with this mobile ECG device and decrease the overall time required for the pharmacist.Together, these findings suggest several strengths and challenges for continued CP efforts in this area.…”