Sommermeyer Dirk, Zou Ding, Ficker Joachim H, Randerath Winfried, Fischer Christoph, Penzel Thomas, Sanner Bernd, Hedner Jan, Grote Ludger
Department of Internal Medicine and Clinical Nutrition, Center for Sleep and Vigilance Disorders, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden.
Institut für Assistenzsysteme und Qualifizierung e.V., Karlsruhe, Germany.
Med Biol Eng Comput. 2016 Jul;54(7):1111-21. doi: 10.1007/s11517-015-1410-8. Epub 2015 Nov 4.
Cardiovascular disease is the main cause of death in Europe, and early detection of increased cardiovascular risk (CR) is of clinical importance. Pulse wave analysis based on pulse oximetry has proven useful for the recognition of increased CR. The current study provides a detailed description of the pulse wave analysis technology and its clinical application. A novel matching pursuit-based feature extraction algorithm was applied for signal decomposition of the overnight photoplethysmographic pulse wave signals obtained by a single-pulse oximeter sensor. The algorithm computes nine parameters (pulse index, SpO2 index, pulse wave amplitude index, respiratory-related pulse oscillations, pulse propagation time, periodic and symmetric desaturations, time under 90 % SpO2, difference between pulse and SpO2 index, and arrhythmia). The technology was applied in 631 patients referred for a sleep study with suspected sleep apnea. The technical failure rate was 1.4 %. Anthropometric data like age and BMI correlated significantly with measures of vascular stiffness and pulse rate variability (PPT and age r = -0.54, p < 0.001, PR and age r = -0.36, p < 0.01). The composite biosignal risk score showed a dose-response relationship with the number of CR factors (p < 0.001) and was further elevated in patients with sleep apnea (AHI ≥ 15n/h; p < 0.001). The developed algorithm extracts meaningful parameters indicative of cardiorespiratory and autonomic nervous system function and dysfunction in patients suspected of SDB.
心血管疾病是欧洲的主要死因，早期检测心血管风险（CR）升高具有临床重要性。基于脉搏血氧饱和度的脉搏波分析已被证明有助于识别CR升高。本研究详细描述了脉搏波分析技术及其临床应用。一种基于匹配追踪的新型特征提取算法被应用于通过单脉搏血氧仪传感器获得的夜间光电容积脉搏波信号的信号分解。该算法计算九个参数（脉搏指数、SpO2指数、脉搏波振幅指数、呼吸相关脉搏振荡、脉搏传播时间、周期性和对称性去饱和、SpO2低于90%的时间、脉搏与SpO2指数之差以及心律失常）。该技术应用于631名因疑似睡眠呼吸暂停而接受睡眠研究的患者。技术故障率为1.4%。年龄和BMI等人体测量数据与血管僵硬度和脉搏率变异性测量值显著相关（PPT与年龄r = -0.54，p < 0.001，PR与年龄r = -0.36，p < 0.01）。复合生物信号风险评分与CR因素数量呈剂量反应关系（p < 0.001），并且在睡眠呼吸暂停患者（AHI≥15次/小时；p < 0.001）中进一步升高。所开发的算法提取了有意义的参数，这些参数表明疑似睡眠呼吸障碍患者的心肺和自主神经系统功能及功能障碍。
