Detection of special points in a photoplethysmogram
The widespread use of photoplethysmography for assessing the functional state of human peripheral blood circulation is due to the simplicity of obtaining high-quality optical pulse curves over a long period of time. Continuous recording of photoplethysmograms (computer monitoring of peripheral blood circulation) allows you to observe periodic fluctuations in regional vascular tone. This is relevant for non-invasive functional diagnostics of a number of diseases of the vascular system. In connection with the above, the contour analysis of photoplethysmograms is of interest for doctors of functional diagnostics - the definition of special points, a kind of "relief" of the pulse cycle. The origin of the characteristic elements of the relief of photoplethysmograms is due to well-defined functional features of the propagation of the wave of elastic deformation of the arterial vascular wall in various segments of the regional vascular bed during the cardiac cycle. Quantitative analysis of the contour of photoplethysmograms includes determining the amplitude of the maximum of the pulse cycle, the dicrotic notch and the diastolic peak or conditional point diastolic peak, if the diastolic rise is missing.
The development of a reliable algorithm for finding special points of photoplethysmograms is a prerequisite for the widespread introduction of automated contour analysis of photoplethysmograms in clinical practice.
Photoplethysmograms recorded in 31 healthy volunteers were analyzed. All photoplethysmograms were divided into 1665 separate pulse cycles. Special points were defined for each pulse cycle of the photoplethysmogram. The statistics of their location on pulse cycles of different duration are obtained. An approximate relationship between the position of special points and the duration of the pulse cycle were established. By minimizing the standard deviation, it is shown that the obtained dependencies are best approximated by 2nd-order polynomials. The average relative error in determining the positions of singular points was ~1% for the maximum of the photoplethysmogram, ~ 3% for the dicrotic notch, and ~ 5% for the diastolic peak.
File with report