Abstract:Background: Wearable and portable devices with ECG capabilities have the potential to enable broad AI-based screening for clinical disorders, but models trained on single-lead signals are limited by challenges accessing the raw signal data from proprietary systems. Additionally, wearable ECGs have traditionally been limited to diagnosing only a few rhythm disorders. Methods: From 385,601 clinical ECGs performed at Yale during 2015-2021, we extracted lead I data, and simulated wearable PDF outputs from commercial wearable devices. Gold-standard diagnosis statements from cardiologists accompanying these ECGs were processed and used to generate 41 clinical labels. We developed a multilabel model to predict these labels from wearable-adapted PDF outputs using an EfficientNet-B3 convolutional neural network with self-supervised pretraining. The model deployment pipeline was optimized to segment and process PDF outputs across all commercially available devices as inputs to the model (Fig A & B). Results: In a held-out test set of 11,611 ECGs, the model accurately detected a diverse set of disorders. For the rhythm disorders sinus tachycardia, sinus bradycardia, atrial fibrillation, atrial flutter, premature atrial contraction, and premature ventricular contraction, the model had AUROCs of 0.99, 0.98, 0.98, 0.93, 0.96 and 0.95 respectively. For the conduction disorders left bundle branch blocks, right bundle branch blocks, 1 st and 2 nd degree AV blocks, the model had AUROCs of 0.98, 0.97, 0.96 and 0.95, respectively. Finally, for key disorders such as STEMI, LVH, and low limb-lead voltage, the model had AUROCs of 0.81, 0.86, and 0.77, respectively (Fig C). Conclusions: We developed and validated a deep learning model to identify conduction disorders and arrhythmias from PDF image outputs of wearable and handheld single-lead devices. This approach allows for automated and easily accessible insights into wearable and portable ECGs at the point of care.

LSTM-Based ECG Classification for Continuous Monitoring on Personal Wearable Devices

High-Performance Personalized Heartbeat Classification Model for Long-Term ECG Signal

An Automatic Cardiac Arrhythmia Classification System with Wearable Electrocardiogram

A Neural Network-Based ECG Classification Processor with Exploitation of Heartbeat Similarity

Wearable Electrocardiogram Quality Assessment Using Wavelet Scattering and LSTM

Patient-Specific ECG Classification with Integrated Long Short-Term Memory and Convolutional Neural Networks

Multiple Functional ECG Signal is Processing for Wearable Applications of Long-Term Cardiac Monitoring

A Wearable ECG Monitor for Deep Learning Based Real-Time Cardiovascular Disease Detection

A Neuromorphic Processing System for Low-Power Wearable ECG Classification.

Artificial Intelligence-Enabled ECG Algorithm Based on Improved Residual Network for Wearable ECG

EMCNet: Ensemble Multiscale Convolutional Neural Network for Single-Lead ECG Classification in Wearable Devices

Multi-level Binarized LSTM in EEG Classification for Wearable Devices

Homecare-Oriented ECG Diagnosis With Large-Scale Deep Neural Network for Continuous Monitoring on Embedded Devices

Scalable and Upgradable AI for Detected Beat-By-Beat ECG Signals in Smart Health

Accurate ECG Classification Based on Spiking Neural Network and Attentional Mechanism for Real-Time Implementation on Personal Portable Devices

DeepBeat: A multi-task deep learning approach to assess signal quality and arrhythmia detection in wearable devices

Inferring ECG from PPG for Continuous Cardiac Monitoring Using Lightweight Neural Network

Hierarchical deep learning for autonomous multi-label arrhythmia detection and classification on real-world wearable electrocardiogram data

Abstract 19135: Deep Learning Model for Detection of Conduction Disorders and Arrhythmias From PDF Outputs of Commercially Available Wearable Devices

Energy-Efficient Real-Time Heart Monitoring on Edge-Fog-Cloud Internet-of-Medical-Things

Wearable ECG Device and Machine Learning for Heart Monitoring