*3.2. Calibration*

After calculating the geometric constant k, according to Equation (21), a calibration procedure was performed to verify the degree of agreement between the measurements made by the YF-S201 sensor and the reference values of (300 ± 2, 450 ± 3, 600 ± 3, 750 ± 4, 900 ± 5, 1050 ± 6, 1200 ± 6, 1350 ± 7, 1500 <sup>±</sup> 8, 1650 <sup>±</sup> 9 and 1800 <sup>±</sup> 9) <sup>×</sup> 10−<sup>6</sup> m3, provided by a syringe especially used in spirometer calibration procedures. As stated in its manual, the syringe was marked at the points corresponding to volumes of interest, according to Table 1, to perform the calibration. Therefore, it was possible to estimate the systematic error and the range of the random error is expected with 95% of probability, achieving application of the bias correction and collecting information about the uncertainty of the instrument along with its measurement range in future measurements, respectively.

**Table 1.** The relationship between the length of stem and volume provided by the syringe.


The acceptable limit of error in spirometry for Forced Vital Capacity (FVC) and Forced Expiratory Volume (FEV) is 3.5% of full scale [25]. Therefore, in this work, we considered the maximum error ε equal to 60 <sup>×</sup> <sup>10</sup>−<sup>6</sup> <sup>m</sup>3, since it represents the limit of 3.4%, satisfying the spirometric conditions.

Considering a small number of repetitions (10 ≤ *n* ≤ 25), and assuming that the mean of the indications follows an approximately normal distribution, the *t*-Student distribution is used to determine the confidence interval. Due to statistical inference, we have:

$$
\varepsilon = t\_{a/2} \frac{s\_0}{\sqrt{n}}.\tag{23}
$$

Ten random measurements were taken to estimate the standard deviation *s*0, which is approximately equal to 98 <sup>×</sup> 10−<sup>6</sup> m3. For a 95% confidence interval, the significance level <sup>α</sup> is 0.05, so *t*α/2= 2.2. Thus, a total of 13 measurements (*n*) should be performed to ensure the statistical significance of the data according to the sampling rules. According to Student's distribution, we obtained *t* = 2.17, which will be used in subsequent tests [63,64]. The correction is added to the measurements to compensate for the effect of the systematic error. The estimated systematic error corresponds to the average value of the measurement error, i.e., the average of n sensor measurements of the same measurand carried out under repeatability conditions minus the conventional true value of the measurand, provided by a standard or a reference instrument. The correction is equal to the negative of the estimated systematic error [63,64].

The uncertainty of measurement (*U*) defines an interval about the result of a measurement that may be expected. In this work, it represents the symmetric range of values around the average error where the random error is expected with 95% of probability. As the probability distribution of the sensor measurements follows the Normal distribution (according to the Normal Probability Plot with R-square equal to 0.99927), we considered Student's distribution to take into account the difference between the standard deviation of the mean and the experimental standard deviation of the mean [63,64]. Thus, the uncertainty of measurement is:

$$
\mathcal{U} = t\_{0.95} \,\,\mathfrak{u},
\tag{24}
$$

where *t*0.95 is the t-factor from Student's distribution considering 95% of confidence level; and *u* is the Type A standard uncertainty, calculated as the experimental standard deviation of the mean.

In this work, the error curve represents the calibration results. It is formed by the center line, which represents the estimated systematic error; and by the upper and lower limits of the range containing the random errors, i.e., the uncertainty of measurement.

### *3.3. Spirometric Tests*

The procedures required to perform rescue ventilation in the practice of CPR must follow the parameters of the American Heart Association [65], which establishes a breath every five or six seconds, that is, 10 to 12 breaths during each 60 s. Approximately 500 <sup>×</sup> 10−<sup>6</sup> m3 of air enters and leaves the lungs of a healthy young adult in a resting state at each respiratory cycle [66]. Therefore, efficient ventilation should provide such a volume of air to the lungs by mouth-to-mouth or using devices for this purpose.

As said before, spirometry is the measure of the air that enters (inspired) and exits (expired) from the lungs. It can be performed during slow breathing or forced expiratory maneuvers. One of the results generated by this technique is an inspired/expired volume versus time graph [25].

Figure 2 shows, as an example, a result of a real spirometry test performed at the Collective Health Laboratory of the Federal University of Ouro Preto, in a 72-year-old male, 58 kg and 1.68 m; following the stress protocol [25], generated by a commercial instrument (Koko brand). The Koko spirometer utilizes a differential pressure sensor, also known as a pneumotachometers, which measures a small (but measurable) pressure difference around a low-value resistance. As the variations in pressure to be detected are small, the material that constitutes the resistance has a high cost. Furthermore, like other commercial spirometers, it cannot be installed on the CPR training dummy because it takes up a lot of space inside it and the response time does not meet the real-time prerequisites for performing CPR training spirometric feedback. It is also worth noting that the spirometer response automatically correlates the measurement range with the patient's breathing conditions, what not desirable during CPR, as the goal is to test for optimal ventilatory maneuvers on a cardiorespiratory arrest victim. For volumes between 300 to 600.10−<sup>6</sup> m3, the typical CPR range, the spirometer has difficulty to perform measurements, as this is not the spirometric assessment range, which is usually around 3 to <sup>6</sup> <sup>×</sup> <sup>10</sup>−<sup>3</sup> m3.

**Figure 2.** Volume versus Time chart generated by Koko spirometer.

Two parameters obtained from these curves are Forced Vital Capacity (FVC) and Forced Expiratory Volume (FEVt). FVC is measured by asking the individual to breathe out until the total lung capacity and expires as rapidly and intensely as possible in a spirometer (Figure 2, FVC = 3 <sup>×</sup> 10−<sup>3</sup> m3). FEVt can be measured in the FVC maneuver at predefined intervals. In the blue line of Figure 2, FEV is approximately 2.5 <sup>×</sup> <sup>10</sup>−<sup>3</sup> <sup>m</sup><sup>3</sup> for 1 s, 2.8 <sup>×</sup> <sup>10</sup>−<sup>3</sup> m3 for 2 s, 2.9 <sup>×</sup> <sup>10</sup>−<sup>3</sup> m3 for 3 s, and practically equal to FVC for 4 s. Besides, the FEV value for 1 s should be approximately 80% of the FVC value [25].

From the blue curve of Figure 2, another parameter is obtained: The Medium Forced Expiratory Flow (FEF25–75%). To calculate FEF25–75%, we mark the points at which 25% and 75% of the FVC were expired on the volume-time curve. A straight line connecting these points is drawn with a duration of 1 s. The vertical distance between the intersection points is FEF25–75% [25].

After calibration, both the YF-S201 sensor and the Koko spirometer were used in a spirometric test, which consisted of applying known air volumes using the syringe: (300 ± 2, 450 ± 3, 600 ± 3, 750 <sup>±</sup> 4, 900 <sup>±</sup> 5, 1050 <sup>±</sup> 6, 1200 <sup>±</sup> 6, 1350 <sup>±</sup> 7, 1500 <sup>±</sup> 8, 1650 <sup>±</sup> 9 and 1800 <sup>±</sup> 9) <sup>×</sup> 10−<sup>6</sup> m3. The total volume of air inside the syringe was passed through the spirometer, lifting the curves from the test. Such curves correspond to the volume of the syringe, considering the measurement error.

Beyond the YF-S201 sensor, Figure 3a,b shows the other components used to perform this test: the syringe outlet and the Koko spirometer, respectively. The spirometer shows the uncertainty of 3% or 100 <sup>×</sup> 10−<sup>6</sup> m3, reproducibility of 0.5% or 150 <sup>×</sup> 10−<sup>6</sup> m3, volume range of 16 <sup>×</sup> 10−<sup>3</sup> m3, flow rate <sup>16</sup> <sup>×</sup> <sup>10</sup>−<sup>3</sup> <sup>m</sup>3/s, and resistance less than 147.1 <sup>×</sup> 103 Pa/(m3s) with the filter.

**Figure 3.** Components for calibrating the spirometer: (**a**) Calibration syringe; (**b**) Koko flow spirometer model 313105.

The measurement results presented by the Koko spirometer were used to verify the quality of the measurements obtained by the system developed in this work, under the same experimental conditions.
