122
H. Schmid-Schönbein
photoplethysmographic activity of severe gingivitis (v.i.). While it is currently unsettled whether this rhythm reflects mechanical influences on the venous outflow or a
neuronally mediated influences (Mück–Weymann), it is clear that respiration-related
activity (RRA) has to be considered as a putatively powerful determinant of the fluctuations seen in the perfusion dynamics the superficial tissues amenable non-invasive
techniques.
These insights led to a strategy in which we strive to reduce the complexity of
the actual measuring devices sensed by patients and volunteers, while the boundary
conditions for measurements have to be closely standardised. This can be done by
placing the subjects into thermocontrolled rooms (with ambient temperature ranging
between 15 °C and 37 °C and by constructing a device allowing to place the hand at the
heart level, well below and well above the heart level. Furthermore, a simple reflection
photoplethysmograph was connected to a battery heated plate (3 cm in diameter),
thus allowing to derive photoplethysmographic data from a vascular region devoid of
vasomotor activity (heating above 31 °C leading to local vasorelaxation and thus to
a situation where passive reaction to changes in the arteriovenous pressure gradient
determine the blood content and thence the photoplethysmographic signal (and its
temporal variation). As described in detail before the data obtained from the forehead
reflection photoplethysmograph, those from the LDA device (we mostly used the
Periflux Models 301 without filter) during measuring periods between 4.5 min (in
the experiments on healthy and diseased gingiva) and 30 min (in measurements on
the volar finger microcirculation.
The data from the PPG of the glabella and the volar finger, as well as those
obtained in the LDA systems used without filters were digitised on line and were
fed into task adapted PC where a dual record was constructed by a software of own
design (Blazek and Hülsbusch). Thus, from each measuring site, two complementary
records were obtained, namely a “double plot” with compressed time series showing
the gross appearance of the fluctuations in amplitudes and a Fourier transform. The
information concerning frequency distribution was obtained by detection of discrete
frequency bands based on the evaluation of the data for intervals between two minutes
and the whole measuring interval (5 min–30 min). In light of the fact that the most
important discrete activity bands were uncovered in the range between 0.25 Hz
(respiration related, about 15/min) and 0.01 (various highly different activities in the
range of 1/min–3/min), the data were initially displayed in such a fashion, that the
peaks could be seen as a function of the logarithm of the frequency. As mentioned
before, these plots not only proved to be far superior to power spectra for the work in
cardiovascular physiology but opened an entirely new “venue” to the comprehension
of complex time series. We propose to call these plots “frequency chromatograms”,
and their application was later extended into the construction of “time frequency
plots” (v.i. Fig. 7.10 in the section Outlook) which greatly aided in the comparison
of various types of low frequency activities in the fluctuation of the effectors of the
autonomous nervous and/or ventilatory system. As an added dividend, the strategem
of “normalising” all frequency chromatograms to the power of the fluctuation in
the ca 1 Hz band (cardiac activity in adult subjects during physical rest) allows to
H. Schmid-Schönbein
photoplethysmographic activity of severe gingivitis (v.i.). While it is currently unsettled whether this rhythm reflects mechanical influences on the venous outflow or a
neuronally mediated influences (Mück–Weymann), it is clear that respiration-related
activity (RRA) has to be considered as a putatively powerful determinant of the fluctuations seen in the perfusion dynamics the superficial tissues amenable non-invasive
techniques.
These insights led to a strategy in which we strive to reduce the complexity of
the actual measuring devices sensed by patients and volunteers, while the boundary
conditions for measurements have to be closely standardised. This can be done by
placing the subjects into thermocontrolled rooms (with ambient temperature ranging
between 15 °C and 37 °C and by constructing a device allowing to place the hand at the
heart level, well below and well above the heart level. Furthermore, a simple reflection
photoplethysmograph was connected to a battery heated plate (3 cm in diameter),
thus allowing to derive photoplethysmographic data from a vascular region devoid of
vasomotor activity (heating above 31 °C leading to local vasorelaxation and thus to
a situation where passive reaction to changes in the arteriovenous pressure gradient
determine the blood content and thence the photoplethysmographic signal (and its
temporal variation). As described in detail before the data obtained from the forehead
reflection photoplethysmograph, those from the LDA device (we mostly used the
Periflux Models 301 without filter) during measuring periods between 4.5 min (in
the experiments on healthy and diseased gingiva) and 30 min (in measurements on
the volar finger microcirculation.
The data from the PPG of the glabella and the volar finger, as well as those
obtained in the LDA systems used without filters were digitised on line and were
fed into task adapted PC where a dual record was constructed by a software of own
design (Blazek and Hülsbusch). Thus, from each measuring site, two complementary
records were obtained, namely a “double plot” with compressed time series showing
the gross appearance of the fluctuations in amplitudes and a Fourier transform. The
information concerning frequency distribution was obtained by detection of discrete
frequency bands based on the evaluation of the data for intervals between two minutes
and the whole measuring interval (5 min–30 min). In light of the fact that the most
important discrete activity bands were uncovered in the range between 0.25 Hz
(respiration related, about 15/min) and 0.01 (various highly different activities in the
range of 1/min–3/min), the data were initially displayed in such a fashion, that the
peaks could be seen as a function of the logarithm of the frequency. As mentioned
before, these plots not only proved to be far superior to power spectra for the work in
cardiovascular physiology but opened an entirely new “venue” to the comprehension
of complex time series. We propose to call these plots “frequency chromatograms”,
and their application was later extended into the construction of “time frequency
plots” (v.i. Fig. 7.10 in the section Outlook) which greatly aided in the comparison
of various types of low frequency activities in the fluctuation of the effectors of the
autonomous nervous and/or ventilatory system. As an added dividend, the strategem
of “normalising” all frequency chromatograms to the power of the fluctuation in
the ca 1 Hz band (cardiac activity in adult subjects during physical rest) allows to