arXiv 2505.075332025-05-12cs.CVcs.AI

IKrNet：在生理变异性中检测心电图中特定药物诱导模式的神经网络

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

IRD, Sorbonne Université, Unité de Modélisation Mathématique et Informatique des Systèmes Complexes, UMMISCO（IRD，索邦大学，数学与复杂系统建模单元，UMMISCO）；University of Evry - Paris-Saclay, IBISC Laboratory（埃弗里-巴黎萨克雷大学，IBISC实验室）；Clinical Investigation Center Paris-Est, CIC-1901, INSERM, Department of Pharmacology Pitié-Salpêtrière University Hospital, Sorbonne Université（巴黎-伊夫里临床研究中心，CIC-1901，INSERM，皮蒂埃-萨尔佩特里耶大学医院药理学部，索邦大学）；Department of Medicine, Vanderbilt University Medical Center（医学系，范德比尔特大学医学中心）；Sorbonne Université, INSERM, Nutrition et Obesities（索邦大学，INSERM，营养与肥胖；系统方法，NutriOmique，AP-HP皮蒂埃-萨尔佩特里耶医院）；systemic approaches, NutriOmique, AP-HP Hôpital Pitié-Salpêtrière

Ahmad Fall, Federica Granese, Alex Lence, Dominique Fourer, Blaise Hanczar, Joe-Elie Salem, Jean-Daniel Zucker, Edi Prifti

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

提出IKrNet神经网络，通过多尺度卷积骨干与双向LSTM融合空间和时间特征，在心率变异性干扰下识别药物诱导的ECG模式，并在990名服用80mg索他洛尔的健康志愿者数据中验证其优于现有模型。

摘要：

即使在包括体力活动、药物和压力影响在内的不同生理条件下，监测和分析心电图（ECG）信号对于准确评估心脏健康也至关重要。然而，当前基于AI的方法往往未能说明这些因素如何相互作用并改变ECG模式，最终限制了其在真实环境中的适用性。本研究提出IKrNet，这是一种新颖的神经网络模型，可在特定生理条件下识别ECG中的药物特异性模式。IKrNet的架构融合了空间和时间动态：使用具有不同感受野大小的卷积骨干来捕获空间特征，同时采用双向长短期记忆模块对时间依赖性进行建模。通过将心率变异性视为生理波动的替代指标，我们在多种场景下评估了IKrNet的性能，包括身体压力条件、仅服药条件以及无药物存在的基线条件。我们的评估遵循一项临床方案，其中990名健康志愿者服用80mg索他洛尔；已知该药物是尖端扭转型室速（一种危及生命的心律失常）的前兆。我们表明，IKrNet在不同生理条件下的准确性和稳定性优于最先进模型，突显了其临床可行性。

英文摘要：

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.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。