Novel Mapping Algorithms for Atrial Fibrillation

心房颤动的新型映射算法

• 批准号: 6836348
• 负责人: Spencer J Melby
• 金额: $ 4.73万
• 依托单位: WASHINGTON UNIVERSITY
• 依托单位国家: 美国
• 财政年份: 2004
• 资助国家: 美国
• 起止时间: 2004-09-01 至 2006-08-31
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/6836348
• 关键词: atrial fibrillation; atrium; clinical research; dogs; epicardial mapping; heart surgery; human subject; infrared spectrometry; interferometry; mathematical model; myocardial ischemia /hypoxia; postdoctoral investigator

DESCRIPTION (provided by applicant): The present surgical treatment for AF involves creating a myriad of incisions in the atrium to block the multiple macroreentrant circuits that were once thought to be the underlying mechanism. However, recently it has been shown that AF is often maintained by a single or small number of spatially stable focal sources. Unfortunately, present technologies of epicardial mapping are time-consuming and do not allow for a map-guided approach for surgical treatment. We propose to develop an algorithm to identify these unique substrates of AF in real time. Using information in a database of prior AF recordings, we will first evaluate AF using the Short Time Fourier transform, which allows rapid three-dimensional representation of a dominant frequency. The development of this algorithm will allow for a better understanding of the mechanisms of AF. After development of the algorithm, we will demonstrate the ability to map in real-time using a canine model of atrial fibrillation. We will next record data from human patients, and compare real-time analysis with off-line analysis to ensure accuracy (wave analysis will be weighed against the more time-consuming traditional manual editing technique). Finally, use of this real-time algorithm and the understanding gained of AF will allow for the development of a novel surgical approach using map-directed treatment.

描述（由申请人提供）：目前的房颤手术治疗涉及在心房内创建无数切口，以阻断曾被认为是潜在机制的多个大折返回路。然而，最近已经表明，AF通常由单个或少量空间稳定的焦点源维持。不幸的是，目前的心外膜标测技术是耗时的，并且不允许用于外科治疗的标测引导方法。我们建议开发一种算法来识别这些独特的基板AF在真实的时间。使用先前AF记录数据库中的信息，我们将首先使用短时傅立叶变换来评估AF，该变换允许主频率的快速三维表示。该算法的发展将允许更好地了解AF的机制。算法开发后，我们将展示使用犬房颤模型实时映射的能力。接下来，我们将记录来自人类患者的数据，并将实时分析与离线分析进行比较，以确保准确性（波形分析将与更耗时的传统手动编辑技术进行权衡）。最后，使用这种实时算法和对AF的理解将允许开发一种使用标测指导治疗的新型手术方法。