Automatic Annotation of Seismocardiogram With High-Frequency Precordial Accelerations

Abstract

Seismocardiogram (SCG) is the low-frequency vibrations signal recorded from the chest using accelerometers. Peaks on dorsoventral and sternal SCG correspond to specific cardiac events. Prior research work has shown the potential of extracting such peaks for various types of monitoring and diagnosis applications. However, annotation of these peaks is not a trivial task and complicated in some subjects. In this paper, an automated method is proposed to annotate these peaks. The high-frequency accelerations obtained from the same accelerometer, used to record SCG with, were used to facilitate the annotation of the SCG. Algorithms were developed for detection of isovolumic moment (IM) and aortic valve closure (AC) points of SCG. Four different envelope calculation methods were used: cardiac sound characteristic waveform (CSCW), Shannon, absolute, and Hilbert. The algorithms were evaluated based on a dataset including 18 subjects undergoing lower body negative pressure and were further tested with another dataset, which included 67 subjects. These datasets had been previously manually annotated. The algorithm based on CSCW envelope calculation produced the highest detection accuracy for both IM and AC. The overall CSCW algorithm detection accuracy for the test dataset was 98.7% and 99.1% for the IM and AC points, respectively.

Automatic Identification of Systolic Time Intervals in Seismocardiogram

“…Recently, an attempt was made to automate the annotation process with envelope based method23. However, the peaks in the SCG signal are searched backward from the corresponding envelope peak which may limit the performance in the presence of additional peak(s) due to artifacts.…”

“…A 5 th order Butterworth filter with the passband of [1 35] Hz2340 and [1 100] Hz41 was employed for SCG and ECG respectively. The same filter orders were used for both SCG and ECG signals to avoid discrepancy in the latencies between the two signals due to filtering.…”

“…1(b). The peaks in the SCG signal correspond to opening and closing of aortic (AO/AC) and mitral (MO/MC) valve2122, while the IM point occurs during the period of rapid change in ventricular pressure23. The intervals PEP, LVET and QS2 illustrated in Fig.…”

Airborne ultrasound surface motion camera: Application to seismocardiography

“…With a few exceptions, these methods have been limited to single-point measurements on the chest surface. While automatic annotation algorithms have been developed [3,4], and published studies demonstrate potential uses for clinical applications [5][6][7][8][9], SCG has so far largely remains confined to scientific laboratories. This is in part due to the lack of physiological and physical understanding of the processes that underlie the phenomena, causing lasting uncertainties as to the exact interpretation of the signal's peaks and valleys, as well as some variability in accelerometer placement.…”

Moving toward automatic and standalone delineation of seismocardiogram signal

“…In contrast, the window selection for the diastolic AC peak is calculated based on the heart rate. The window selection (high confidence occurrence region), especially for AC, is very important since many adjacent peaks can have similar morphology and hence lead to misdetection [7].…”