The results obtained from this study can be employed to enhance the extraction of clinically valuable information such as systolic time intervals.

No abstract

a b s t r a c tThis study presents an evaluation of an advanced Doppler radar-based method for detection of vital signs, presence, and activity of a human subject in a test room with radar-signal reflecting aluminum-coated surfaces. Ten test subjects lay in four positions, and they sat in two locations in the room, both breathing normally and holding their breath. The mean ratios of the pulse rates determined from the radar signal and electrocardiography and respiration reference signals were 110% (respiration) and 99% (heartbeat), and the mean occupied and empty room radar signal variance ratios were 608 (breathing) and 20 (breath-hold). In a one-subject activity monitoring test, walking, standing and lying activities could be well separated from the radar signal. The results are promising and the proposed system seems to have potential to be used in position-independent health and activity monitoring of, for example, elderly people in care homes or intoxicated people in police custody.

The vibrations produced by the cardiovascular system that are coupled to the precordium can be noninvasively detected using accelerometers. This technique is called seismocardiography. Although clinical applications have been proposed for seismocardiography, the physiology underlying the signal is still not clear. The relationship of seismocardiograms of on the back-to-front axis and cardiac events is fairly well known. However, the 3-D seismocardiograms detectable with modern accelerometers have not been quantified in terms of cardiac cycle events. A major reason for this might be the degree of intersubject variability observed in 3-D seismocardiograms. We present a method to quantify 3-D seismocardiography in terms of cardiac cycle events. First, cardiac cycle events are identified from the seismocardiograms, and then, assigned a number based on the location in which the corresponding event was found. 396 cardiac cycle events from 9 healthy subjects and 120 cardiac cycle events from patients suffering from atrial flutter were analyzed. Despite the weak intersubject correlation of the waveforms (0.05, 0.27, and 0.15 for the x-, y-, and z-axes, respectively), the present method managed to find latent similarities in the seismocardiograms of healthy subjects. We observed that in healthy subjects the distribution of cardiac cycle event coordinates was centered on specific locations. These locations were different in patients with atrial flutter. The results suggest that spatial distribution of seismocardiographic cardiac cycle events might be used to discriminate healthy individuals and those with a failing heart.

Accelerometer-based seismocardiography and sternal acceleration ballistocardiography are promising approaches to the noninvasive detection of precordial vibrations. However, in order to be widely accepted as diagnostic or even prognostic tools, clinical validation and standardization of these methods are necessary. In precordial vibration studies, using all three axes instead of one in cardiac vibration analysis is anticipated to enable more accurate cardiac event detection. Simultaneously acquired electrocardiography, photoplethysmography, and respiration information are considered as promising ways to enhance seismocardiogram analysis. In this article, an easy-to-use system that combines three-dimensional seismocardiography, electrocardiography, photoplethysmography, and respiration measurements is described, and its performance is demonstrated. In the test measurements, the system demonstrated its capability to capture accurate cardiovascular data.