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.

Efficient Detection of Ion Channel Switching through Rapid Random Forest Models

Identifying Ion Channel Genes Related to Cardiomyopathy Using A Novel Decision Forest Strategy

Rapid and non-invasive diagnosis of hyperkalemia in patients with systolic myocardial failure using a model based on machine learning algorithms

AutoMLP: A Framework for the Acceleration of Multi-Layer Perceptron Models on FPGAs for Real-Time Atrial Fibrillation Disease Detection.

Evaluating the predictive accuracy of ion channel models using data from multiple experimental designs

Single shot diagnosis of ion channel dysfunction from assimilation of cell membrane dynamics

A deep learning approach to real-time Markov modeling of ion channel gating

Real-Time Event-Driven Classification Technique for Early Detection and Prevention of Myocardial Infarction on Wearable Systems

Rapidly detecting disorder in rhythmic biological signals: A spectral entropy measure to identify cardiac arrhythmias

Real-time ultra-low power ECG anomaly detection using an event-driven neuromorphic processor

Detection Analysis of Epileptic EEG Using a Novel Random Forest Model Combined with Grid Search Optimization.

Single shot detection of alterations across multiple ionic currents from assimilation of cell membrane dynamics

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

A Delay-based Neuromorphic Processor for Arrhythmias Detection

Energy-Efficient Reservoir Computing Based on Solution-Processed Electrolyte/Ferroelectric Memcapacitive Synapses for Biosignal Classification

Predicting Potential Drug Targets from Ion Channel Proteins Based on Ensemble Learning

Electrocardiogram Analysis Based on Multiscale Fast Sample Entropy and Random Forest

Novel screening techniques for ion channel targeting drugs

Congestive heart failure detection using random forest classifier

Detecting Atrial Fibrillation in Real Time Based on PPG via Two CNNs for Quality Assessment and Detection

A resource-efficient ECG diagnosis model for mobile health devices