Deep learning reduces false heart alarms in intensive care units
Physics-Informed Deep Learning for False Ventricular Tachycardia Alarm Reduction in the ICU
Machine Learning
Summary
False heart alarms in intensive care units cause a lot of unnecessary noise and make it hard for medical staff to focus on real emergencies. The researchers developed a computer method that uses deep learning combined with knowledge about how blood pressure works to better recognize true heart problems. Their approach trains the system to understand realistic blood pressure patterns, which helps it ignore faulty signals in heart monitoring. Tested on a challenging dataset, their method improved accuracy and required fewer labeled heart data examples.
ventricular tachycardiafalse alarmsintensive care unitdeep learningECGWindkessel modelhemodynamicsdata augmentationlatent representationreal-time monitoring
Authors
Athanasios Papastathopoulos-Katsaros, Alexandra Stavrianidi, Zhandong Liu
Abstract
False ventricular tachycardia (VT) alarms are a leading contributor to alarm fatigue in intensive care units. We propose a deep learning framework combining a 1D SE-ResNet with ICU-realistic data augmentations and a physics-informed auxiliary reconstruction task based on the three-element Windkessel hemodynamic model, implemented as a differentiable forward simulation. By requiring the network's latent representation to produce physiologically plausible arterial pressure waveforms, artifact-driven ECG patterns are penalized while true VT remains coherent across modalities. Evaluated on the VTaC benchmark under a strict real-time protocol (10-second pre-alarm window), our method achieves a 5-point Challenge Score improvement over prior state-of-the-art. Ablation studies confirm that the physics-informed objective is the primary performance driver, providing gains in accuracy, 2x label efficiency, and more localized and clinically meaningful ECG segments.