Pulse transit time based respiratory rate estimation with singular spectrum analysis

• Published in: Medical & biological engineering & computing Vol. 58; no. 2; pp. 257 - 266
• Main Authors: Ding, Xiaorong, Yan, Bryan P., Karlen, Walter, Zhang, Yuan-Ting, Tsang, Hon Ki
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.02.2020; Springer Nature B.V
• Subjects: Biomedical and Life Sciences; Biomedical Engineering and Bioengineering; Biomedicine; Blood pressure; Breathing; Computer Applications; EKG; Electrocardiography; Heart rate; Human Physiology; Imaging; Original Article; Parameter estimation; Power spectral density; Radiology; Respiration; Respiratory rate; Spectrum analysis; Transit time; Singular spectrum analysis; Respiratory rate; Blood pressure; Pulse transit time
• ISSN: 0140-0118; 1741-0444; 1741-0444
• DOI: 10.1007/s11517-019-02088-6

Respiratory rate (RR) is an important vital sign which can be difficult to measure accurately and unobtrusively in routine clinical practice. Pulse transit time (PTT), on the other hand, is unobtrusive to collect from electrocardiogram (ECG) and photoplethysmogram (PPG) signals. Using PTT is a novel method to estimate and monitor blood pressure (BP) and RR. This study aimed to estimate continuous RR using PTT with singular spectrum analysis to extract respiratory components. The performance of this method was validated on 17 subjects who carried out spontaneous breathing and controlled deep breathing conditions. Three types of estimated RR parameters (average RR by power spectral density (PSD) (RR PSD ), number of breaths (RR # ), and instantaneous RR (RR inst )) were compared with the corresponding reference RR. The reference RR was collected using a respiratory belt. Our findings demonstrate that the PTT signal reliably tracked respiratory variation with a root mean square error of 0.84, 1.11, and 0.74 breaths/min for RR PSD , RR # , and RR inst estimations, respectively. Overall, RR estimated by PTT was more accurate than heart/pulse rate interval, QRS area, and PPG amplitude, which were also extracted from ECG and PPG. The results suggest that it may be feasible to use PTT as an estimation of RR and that ECG and PPG may be relied upon for monitoring not only RR but also BP and heart rate. Graphical abstract The Pulse Transit Time (PTT) based Respiratory Rate (RR) estimation with Singular Spectrum Analysis (SSA) provides a superior performance than the method with other respiratory indicators extracted from Electrocardiogram (ECG) or Photoplethysmogram (PPG)