IKrNet: A Neural Network for Detecting Specific Drug-Induced Patterns in Electrocardiograms Amidst Physiological Variability

2025-05-12

AI Summary

Current AI methods for ECG analysis often neglect confounding physiological conditions—such as physical activity, pharmacological interventions, and stress—that distort signal morphology, thereby limiting clinical generalizability. To address this, we propose IKrNet, a physiology-aware neural network integrating multi-scale convolutional layers with bidirectional LSTMs. Crucially, it is the first to explicitly incorporate heart rate variability (HRV) as a proxy metric for underlying physiological fluctuations and introduces a physiology-variation-aware training paradigm. Evaluated on a clinical cohort of 990 subjects undergoing pharmacological challenge tests (e.g., sotalol administration), IKrNet demonstrates significant performance gains over state-of-the-art methods across diverse physiological scenarios. It achieves robust, drug-specific ECG pattern recognition—particularly QT-interval prolongation—and markedly improves the reliability and clinical feasibility of early torsades-de-pointes risk stratification.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

Monitoring and analyzing electrocardiogram (ECG) signals, even under varying physiological conditions, including those influenced by physical activity, drugs and stress, is crucial to accurately assess cardiac health. However, current AI-based methods often fail to account for how these factors interact and alter ECG patterns, ultimately limiting their applicability in real-world settings. This study introduces IKrNet, a novel neural network model, which identifies drug-specific patterns in ECGs amidst certain physiological conditions. IKrNet's architecture incorporates spatial and temporal dynamics by using a convolutional backbone with varying receptive field size to capture spatial features. A bi-directional Long Short-Term Memory module is also employed to model temporal dependencies. By treating heart rate variability as a surrogate for physiological fluctuations, we evaluated IKrNet's performance across diverse scenarios, including conditions with physical stress, drug intake alone, and a baseline without drug presence. Our assessment follows a clinical protocol in which 990 healthy volunteers were administered 80mg of Sotalol, a drug which is known to be a precursor to Torsades-de-Pointes, a life-threatening arrhythmia. We show that IKrNet outperforms state-of-the-art models' accuracy and stability in varying physiological conditions, underscoring its clinical viability.

Problem

• Detects drug-induced ECG patterns despite physiological variability
• Improves accuracy in varying conditions like stress and drug intake
• Addresses limitations of current AI methods in real-world ECG analysis

Innovation

• Neural network detects drug-induced ECG patterns
• Combines convolutional and bi-directional LSTM layers
• Uses heart rate variability for physiological context