Advanced Signal Processing Methods for Analysis of Fibrillatory Waves

几十年来，信号处理方法一直用于从人类心脏的心电图记录中提取关键信息，用于诊断包括心房颤动（AF）在内的许多心脏病。房颤是最常见的心律失常，仅在北美就有超过3400万人受到影响，每年的医疗费用高达60亿美元。不幸的是，由于疾病的复杂性和动态性，目前的处理方法在针对房颤来源方面取得了有限的成功。******用于AF诊断的传统处理方法基于顺序数据收集，通常利用来自单个记录通道的局部信息来定位AF源，而不考虑通道之间的时间关联来提取波的传播特征。这可以归因于这样一个事实，即从这些系统中获得的记录具有较低的空间分辨率，并且由于顺序采集，心房中颤动波的全局快照是不可行的。最近的努力集中在使用替代记录设备来创建心房内纤颤波传播的全景视图。然而，这些研究报告的结果往往是相互矛盾的。此外，这些记录设备在心脏小腔内的放置和操作的高成本和困难阻碍了这些新的诊断系统的广泛使用。******我的研究计划的目的是开发新的信号处理方法，利用从传统记录系统获得的数据来研究人类心脏中的纤颤波传播。这将通过开发准确估计局部激活间隔的方法和通过分析同时记录的信号中局部活动间隔的时间关联来获得与局部信息相反的区域信息来实现。虽然区域信息不能提供心房内的全局波传播图，但它具有未开发的潜力，可以揭示心房内纤颤波轨迹的足迹和区域的潜在生理特性。我们的处理方法将通过从详细的3D模型中生成逼真的模拟数据来开发和验证。我们的方法在临床数据库中的适用性也将被研究。为了实现这个研究计划的目标，将有5名研究生在我的指导下进行一些项目。******该研究项目的成功结果将提供一种独特的方法来研究和更好地理解使用模拟/真实序列数据的纤颤波的传播，并将促进后续研究以验证人体数据的发现。这可以缩短临床程序，减轻加拿大卫生保健系统的经济负担，并帮助数百万生活受房颤影响的人。

Signal processing methods have been used for decades to extract critical information from electrogram recordings of the human heart for diagnosis of numerous heart diseases including atrial fibrillation (AF). AF is the most common cardiac arrhythmia affecting more than 34 million people with an annual cost of $6 billion just in North America. Unfortunately, the current processing methods have had limited success in targeting the sources of AF due to the complex and dynamic nature of the disease. ******The conventional processing methods used for AF diagnosis are based on sequential data collection and often utilize the local information from individual recording channels to localize AF sources without considering temporal association between the channels to extract wave propagation characteristics. This can be attributed to the fact that the recordings obtained from these systems have low spatial resolution and, due to the sequential acquisition, a global snapshot of fibrillatory waves in atria is not feasible. Recent efforts focus on employing alternative recording equipment to create a panoramic view of the fibrillatory wave propagation in atria. However, the reported outcomes of these studies have been often contradictory. In addition, the high cost and difficulty in placement and maneuvering of these recording devices into small chambers of the heart have prohibited the widespread use of these new diagnostic systems. ******The aim of my research program is to develop new signal processing methods to study fibrillatory wave propagation in the human heart using data obtained from the conventional recording systems. This will be achieved by developing methods for accurate estimation of local activation intervals and by analyzing the temporal association of local active intervals among simultaneously recorded signals to obtain regional information, in contrast to local information. Although the regional information cannot provide a global wave propagation map in atria, it has an unexplored potential to reveal footprints of the trajectory of the fibrillatory waves and the underlying physiologic properties of the regions within the atria. Our processing methods will be developed and validated by generating realistic simulated data from a detailed 3D model. The applicability of our methods on a clinical database will also be studied. To achieve the goals of this research program, a number of projects will be conducted by five graduate students under my supervision.******Successful results from this research program will provide a unique approach to investigate and better understand the propagation of fibrillatory waves using simulated/real sequential data and will facilitate a subsequent study to validate the findings on human data. This can lead to shorter clinical procedures, reduce the financial burden on the Canadian health care system and help the millions of people whose lives are affected by AF.