![[2602.20165] VISION-ICE: Video-based Interpretation and Spatial Identification of Arrhythmia Origins via Neural Networks in Intracardiac Echocardiography](https://arxiv.org/static/browse/0.3.4/images/arxiv-logo-fb.png){kind=logo}
[2602.20165] VISION-ICE: Video-based Interpretation and Spatial Identification of Arrhythmia Origins via Neural Networks in Intracardiac Echocardiography
Summary
The paper presents VISION-ICE, an AI framework utilizing intracardiac echocardiography to identify arrhythmia origins, achieving 66.2% accuracy in a classification task.
Why It Matters
This research highlights the potential of AI in enhancing the efficiency of arrhythmia localization, which could lead to quicker and more accurate electrophysiological interventions. By leveraging ICE, the study addresses the limitations of traditional mapping techniques, making it significant for both clinical practice and technological advancement in cardiology.
Key Takeaways
- VISION-ICE uses AI to analyze intracardiac echocardiography for arrhythmia localization.
- The model achieved a mean accuracy of 66.2%, significantly better than a random baseline.
- This approach could reduce procedural times in electrophysiological interventions.
- Future work aims to expand the dataset for improved model robustness.
- The study underscores the clinical promise of integrating deep learning with echocardiographic imaging.
Computer Science > Computer Vision and Pattern Recognition arXiv:2602.20165 (cs) [Submitted on 4 Feb 2026] Title:VISION-ICE: Video-based Interpretation and Spatial Identification of Arrhythmia Origins via Neural Networks in Intracardiac Echocardiography Authors:Dorsa EPMoghaddam, Feng Gao, Drew Bernard, Kavya Sinha, Mehdi Razavi, Behnaam Aazhang View a PDF of the paper titled VISION-ICE: Video-based Interpretation and Spatial Identification of Arrhythmia Origins via Neural Networks in Intracardiac Echocardiography, by Dorsa EPMoghaddam and 5 other authors View PDF HTML (experimental) Abstract:Contemporary high-density mapping techniques and preoperative CT/MRI remain time and resource intensive in localizing arrhythmias. AI has been validated as a clinical decision aid in providing accurate, rapid real-time analysis of echocardiographic images. Building on this, we propose an AI-enabled framework that leverages intracardiac echocardiography (ICE), a routine part of electrophysiology procedures, to guide clinicians toward areas of arrhythmogenesis and potentially reduce procedural time. Arrhythmia source localization is formulated as a three-class classification task, distinguishing normal sinus rhythm, left-sided, and right-sided arrhythmias, based on ICE video data. We developed a 3D Convolutional Neural Network trained to discriminate among the three aforementioned classes. In ten-fold cross-validation, the model achieved a mean accuracy of 66.2% when evaluated on four p...
