Automated Detection of Acute Myocardial Infarction Using Asynchronous Electrocardiogram Signals—Preview of Implementing Artificial Intelligence With Multichannel Electrocardiographs Obtained From Smartwatches: Retrospective Study

Abstract

Background When using a smartwatch to obtain electrocardiogram (ECG) signals from multiple leads, the device has to be placed on different parts of the body sequentially. The ECG signals measured from different leads are asynchronous. Artificial intelligence (AI) models for asynchronous ECG signals have barely been explored. Objective We aimed to develop an AI model for detecting acute myocardial infarction using asynchronous ECGs and compare its performance with that of the automatic ECG interpretations provided by a commercial ECG analysis software. We sought to evaluate the feasibility of implementing multiple lead–based AI-enabled ECG algorithms on smartwatches. Moreover, we aimed to determine the optimal number of leads for sufficient diagnostic power. Methods We extracted ECGs recorded within 24 hours from each visit to the emergency room of Ajou University Medical Center between June 1994 and January 2018 from patients aged 20 years or older. The ECGs were labeled on the basis of whether a diagnostic code corresponding to acute myocardial infarction was entered. We derived asynchronous ECG lead sets from standard 12-lead ECG reports and simulated a situation similar to the sequential recording of ECG leads via smartwatches. We constructed an AI model based on residual networks and self-attention mechanisms by randomly masking each lead channel during the training phase and then testing the model using various targeting lead sets with the remaining lead channels masked. Results The performance of lead sets with 3 or more leads compared favorably with that of the automatic ECG interpretations provided by a commercial ECG analysis software, with 8.1%-13.9% gain in sensitivity when the specificity was matched. Our results indicate that multiple lead-based AI-enabled ECG algorithms can be implemented on smartwatches. Model performance generally increased as the number of leads increased (12-lead sets: area under the receiver operating characteristic curve [AUROC] 0.880; 4-lead sets: AUROC 0.858, SD 0.008; 3-lead sets: AUROC 0.845, SD 0.011; 2-lead sets: AUROC 0.813, SD 0.018; single-lead sets: AUROC 0.768, SD 0.001). Considering the short amount of time needed to measure additional leads, measuring at least 3 leads—ideally more than 4 leads—is necessary for minimizing the risk of failing to detect acute myocardial infarction occurring in a certain spatial location or direction. Conclusions By developing an AI model for detecting acute myocardial infarction with asynchronous ECG lead sets, we demonstrated the feasibility of multiple lead-based AI-enabled ECG algorithms on smartwatches for automated diagnosis of cardiac disorders. We also demonstrated the necessity of measuring at least 3 leads for accurate detection. Our results can be used as reference for the development of other AI models using sequentially measured asynchronous ECG leads via smartwatches for detecting various cardiac disorders.

Comparison of Apple Watch vs KardiaMobile: A Tale of Two Devices

“…In contrast, KM was more accurate for AF detection and can be attributed to differences in the artificial intelligence programs used by both devices. Each program has characteristics J o u r n a l P r e -p r o o f that allow its respective device to achieve a higher degree of accuracy in a particular heart rhythm, due to differences such as ECG datasets (15). Another possible explanation for the AW4's lower AF rhythm detection is that its artificial intelligence misinterpreted certain rhythm variations ( 16) such as, in the case of two patients, the Apple Watch recorded atrial fibrillation (AF) that was later confirmed to be premature atrial complexes (PACs).…”

Artificial Intelligence-Enabled ECG Algorithm for the Prediction of Coronary Artery Calcification

“…The ECG is a ubiquitous and standardized tool in clinical medicine that reflects physiological and structural condition of the heart and also give valuable diagnostic clues for systemic conditions ( 19 ). Recent AI techniques have been applied to ECGs for the automatic classification or diagnosis of various cardiac diseases, such as arrhythmia and ischemia ( 28 – 31 ). Moreover, there is growing evidence that advanced AI techniques with deep convolutional neural networks are capable of detecting subtle signals and patterns from ECGs that do not fit traditional knowledge and are unrecognizable by the human eye ( 19 ); for example, an AI-enabled ECG algorithm has been shown to be capable of identifying patients with atrial fibrillation during normal sinus rhythm, which has important implications for atrial fibrillation screening and the management of patients with unexplained stroke ( 28 ); another AI-enabled ECG algorithm has been shown to be capable of predicting 1-year all-cause mortality even within a subset of patients with ECGs interpreted as normal by a physician ( 32 ).…”

Using the Apple Watch to Record Multiple-Lead Electrocardiograms in Detecting Myocardial Infarction: Where Are We Now?

“…In a retrospective study of the use of artificial intelligence simulation, 3 asynchronous ECG leads (I, II, and V 5 ) were able to detect MI, and 4 leads (I, II, V 2 , and V 5 ) were an even better predictor. 30 Although no studies based on these conclusions have been performed in humans, these results provide a potential direction for an Apple Watch study. Developing an easier self-check protocol that requires only 3 or 4 recordings may help regular users avoid mistakes and save time.…”