Instrumentation & Measurement Magazine 23-4 - 85

HYT 271 and the ultrafast
Hsensor under these conditions are compared, both
in terms of response times
and then of step response
versus time.
Regarding the influence
of the air flux velocity, it can
be appreciated how sensor
response times decreased
with the increase of the
flow velocity. The higher
variability observed in
some of the measurements
can be addressed to a not
perfectly static positioning of the sensor during
the test.
With respect to the declared value of 10 ms for
1/erise, the fastest response
times appeared higher
probably due to the delay introduced from the
valve opening (around
10 ms). However, considering that in [12] the best
trise and tfall obtained were,
respectively, 1 and 1.8 s,
Fig. 6. Comparison between the error of HYT 271 and of H sensor from BS in measuring the maximum value of a typical
the values obtained here
AH variation during breath, related to the respiratory rate.
with the new H sensor
(< 700 ms) appear significantly lower. These values appear also very interesting if Conclusion
compared with the fastest response times obtained from sim- An improved measuring portable device has been proposed
ilar devices presented in previous literature: 1.4 s in [10] and for real-time monitoring of temperature and humidity during
2 s in [11]. Furthermore, comparing the response times ob- mechanical ventilation. Focusing on the possibility to evaltained during H rise and fall, no significant differences can be uate HME performances with higher accuracy and shorter
observed with the new H sensor, different from what could response times, an accurate in vitro set up has been designed to
be highlighted in the use of HYT 271, which appeared slower test the improvement brought by novel ultrafast humidity and
in its response to decreasing H, thus limiting the possibility temperature commercial sensors in a portable device compatito correctly follow inhalation and exhalation phases during ble with ICU routine. The improved response times (57 ms for
T τrise and around 220 ms for H τrise/fall) allowed the system to estifast respiratory rates.
In order to properly assess the range of respiratory rates mate a reduction of almost 5 times of the error in the detection
monitorable using the new sensor, an analysis was per- of the proper H value if compared with the previous version
formed, similar to what was presented in [12]. Both sensors of the device. Furthermore, both sensors, due to their reduced
were modeled as two first order systems [14] using the time dimension, light-weight and flexibility, could be easily inteconstants (τrise and of τfall at 63.8%) obtained experimentally grated and conditioned inside the portable system, without
with the sensors perpendicular in front of a 45 l/min flux. affecting the overall portability and compatibility with the ICU
Using those models [12], it was possible to quantify the per- routine. Overall, the integration of the characterized sensors
centage of error of the measured H at different respiratory in the portable device allows a real-time and reliable in vivo
rates.
reconstruction of the T and H of the breath in a wide range of
Results obtained (Fig. 6) highlight how the use of the novel respiratory frequencies, thus making it particularly suitable
resistive H sensor can reduce the error in detecting the H of al- for HMEs monitoring in clinical environments. Starting from
most 5 times compared to HYT 271, allowing the system to the results obtained here, which confirmed the possibility to
monitor respiratory rates up to 35 acts per minute with a max- improve the accuracy in evaluating HME performance, fuimum error of 5%.
ture developments will integrate the presented device in ICU
June 2020

IEEE Instrumentation & Measurement Magazine 85

Instrumentation & Measurement Magazine 23-4

Table of Contents for the Digital Edition of Instrumentation & Measurement Magazine 23-4