Motion, Artifact Cancelling MIMO Method for Ambulatory Respiratory Rate Monitorin

描述(申请人提供)：呼吸短促和呼吸困难与心力衰竭(HF)和/或慢性阻塞性肺疾病(COPD)患者的病情恶化直接相关。持续监测这些患者的呼吸状态可以提醒照顾者及早采取干预措施来管理疾病症状，从而防止灾难性事件并提高生活质量。不幸的是，呼吸频率和呼吸模式的持续监测在临床实践中经常被忽视或忽视，因为进行这些测量通常存在困难，特别是在非插管的门诊环境中。目前捕捉患者呼吸道的测量方法是最准确的，但管理困难，患者往往无法忍受，而传统上依赖于捕捉胸部运动的方法由于运动伪影而准确性较差。提出了一种基于MIMO的运动伪影消除多导联阻抗测量方法，用于健壮的动态呼吸频率监测。我们最近证明，在多导联阻抗测量的门诊受试者中，呼吸和运动伪影理论上是可以分离的。与呼吸信号相比，运动伪影是零星的、局部性的高能非平稳事件。由于传统数据采集系统的动态范围和量化误差有限，源分离算法难以提取可靠的呼吸信号信息。所提出的MIMO系统方法通过将统计学习直接集成在模数转换过程中来克服这些限制。因此，这种小型化的硬件实现方法能够连续、实时和省电地跟踪被运动伪影破坏的呼吸信号。基于MIMO算法的初步研究表明，该算法在动态呼吸频率监测中取得了良好的效果。在这项研究中，我们计划开发MIMO系统的优化硬件原型，以验证其在真实移动条件下的性能。我们认为，由于电极的常规使用，基于MIMO的多导联阻抗测量方法很可能被临床医生和患者接受。在未来的工作中，我们设想将拟议的研究与共享同一套电极的心电监测相结合，为普通病房以及门诊患者的家庭健康/健康监测提供连续和准确的心肺信息。

DESCRIPTION (provided by applicant): Shortness of breath and difficulty in breathing are directly associated with deteriorating conditions in patients with heart failure (HF) and/or chroni obstructive pulmonary disease (COPD). Continuous monitoring of respiratory status in these patients can alert caregivers to administer early interventions to manage disease symptoms, thus preventing catastrophic events and improving quality of life. Unfortunately, continuous monitoring of respiratory rate and pattern is often overlooked or neglected in clinical practice due to the general difficulty in performing these measurements, especially in non-intubated ambulatory settings. Present measurement methods that capture the patient's airway are most accurate but difficult to administer and often intolerable for the patient, whereas methods that rely on capturing chest motion historically suffer from poor accuracy due to motion artifacts. We propose a MIMO-based motion artifact cancelling multi-lead impedance measurement method for robust ambulatory respiratory rate monitoring. We have recently demonstrated that respiration and motion artifacts are theoretically separable in ambulatory subjects with multi-lead impedance measurements. The motion artifacts are sporadic and localized high energy non-stationary events compared to respiratory signals. Due to the limited dynamic range and quantization errors of the conventional data acquisition system, source separation algorithms struggle to extract reliable respiratory signal information. The proposed MIMO system approach overcomes these limitations by integrating statistical learning directly within analog to digital conversion process. As a result, this miniaturized hardware realization method enables continuous, real-time and power-efficient tracking of the respiratory signal corrupted by motion artifacts. Preliminary study of the MIMO based algorithm has shown promising results for ambulatory respiratory rate monitoring. In this research we plan to develop the optimized hardware prototype of the MIMO system to validate the performance under real ambulatory conditions. We believe that the MIMO based multi-lead imped- ance measurement method is likely to be accepted by both clinicians and patients due to routine use of electrodes. Going forward, in our future work we envision integrating the proposed research with ECG monitoring sharing same set of electrodes to provide continuous and accurate cardiorespiratory information for general ward as well as for home health/wellness monitoring of ambulatory patients.