The present invention relates to a breathing simulator for calibrating a gas analyzer designed for determining a variable rate of oxygen consumption, usually named V 0 ₂ , and/or a variable rate of carbon dioxide production, usually named V C0 ₂ , these rates applying to a living subject.

This invention also relates to a facility for calibrating such a gas analyzer, this facility including, amongst others, a breathing simulator. This invention also relates to a calibration method for calibrating such a gas analyzer and to a determination method for determining a variable rate of oxygen consumption V0 ₂ and/or a variable rate of carbon dioxide production VC0 ₂ of a living subject with a calibrated gas analyzer.

Indirect calorimetry is known for measuring respiratory gas flows of individuals, in particular oxygen consumption and production of carbon dioxide, which allows evaluating the energetic dependency of a patient. Different measure configurations can be considered, in particular when a patient is lying, at rest, with its head under a canopy.

Indirect calorimetry is implemented with a gas analyzer, also called indirect calorimeter, which allows measuring the oxygen or carbon dioxide concentration in expired gases via different techniques, e.g. paramagnetic or chemical for oxygen or infrared or chemical for carbon dioxide.

A known problem with gas analyzers is that their precision varies along time and that they must be calibrated before being used in order for the values determined with each gas analyzer to be accurate.

Calibration of a gas analyzer can be performed by connecting this gas analyzer to a tank including a known gas mixture with a predetermined composition. Usually, the pressure of the gas mixture leaving this tank is between 3 and 7 bars. Then, the gas mixture under pressure is provided to the gas analyzer and the oxygen concentration measured by the gas analyzer is compared to the oxygen concentration in the known gas mixture. In addition or as a variant, the carbon dioxide concentration measured by the gas analyzer is compared to the carbon dioxide concentration in the known gas mixture.

In the present description and in the appended claims, a concentration of a gas in a gas mixture corresponds to the fraction of this gas in this mixture and is expressed as a percentage.

If one uses the gas mixture directly coming from the tank, its pressure is not representative of the pressure of the gases expired by a patient, these gases being close to atmospheric pressure, which might decrease the accuracy of the concentration measure.

Moreover, when a gas analyzer is to be calibrated, it is necessary to take several measures at different flowrates, e.g. at 5 l/min, 6 l/min, 8 l/min etc. which induces a high consumption of the gas mixture included in the tank. Since calibration of a gas analyzer has to be often done, typically every morning and every afternoon on a day of use of the gas analyzer, the previously described technique induces a large consumption of a known gas mixture, which is expensive and requires handling of a large number of tanks.

It is also known from US-A-4 680 956 to calibrate a respiratory analyzer with a mixture of air and carbon dioxide moved by a reciprocating pump. The gas flows through a bladder and one-way pressure valves, which does not allow precisely controlling its flowrate or changing the proportions in the gas mixture.

This invention aims at solving these problems with a new breathing simulator which allows efficient calibration of a gas analyzer, with use of a little amount of a known gas mixture with a predetermined composition.

To this end, the invention relates to a breathing simulator for calibrating a gas analyzer, this gas analyzer being designed for determining a variable rate of oxygen consumption, or V 0 _2, and/or for determining a variable rate of carbon dioxide production, or V C0 ₂ , of a living subject. This breathing simulator includes at least a first inlet, for connection to a reservoir of a gas mixture with a predetermined composition, a second inlet for connection to the ambient atmosphere and an outlet for connection to the gas analyzer. The breathing simulator also includes a gas mixer for mixing the gas mixture coming from the first inlet with air coming from the second inlet and three fluid lines, namely a first fluid line connecting the first inlet to the gas mixer, a second fluid line connecting the second inlet to the gas mixer and a third fluid line connecting the gas mixer to the outlet. The breathing simulator also includes two flow regulators, namely a first flow regulator mounted on the first fluid line and a second flow regulator mounted on a third fluid line. A metering pump is mounted on the third fluid line and an electronic control unit pilots at least the first and second fluid regulators, via first and second electronic signals, for simulating different breathing regimes.

Thanks to the invention, the two flow regulators respectively mounted on the first and third fluid lines allow modifying the flowrate of combined gas mixture and air provided to the gas analyzer and also to modify the proportion of gas mixture and the proportion of air in this combination. Thus, different breathing regimes can be simulated with the breathing simulator, with relatively low flowrates, which results in a low consumption of gas mixture coming out of the reservoir, as compared to the one necessary for the prior art technique.

According to advantageous but optional aspects of the invention, such a breathing simulator might incorporate one or several of the following features, taken in any technically admissible combination:

- The gas mixer, the first, second and third fluid lines, the first and second regulators, the metering pump and the electronic control unit are enclosed within a box forming a housing for these parts.

- The breathing simulator includes a T joint mounted downstream of the outlet for connecting this outlet, on the one side, to the gas analyzer and, on the other side, to the atmosphere.

- The metering pump has a standard temperature and pressure working capacity between 0,5 and 4 l/min, preferably between 1 ,5 and 3 l/min, still preferably of about 2 l/min

- The metering pump is piloted by the electronic control unit, via a third electronic signal, to work at the same speed, irrespective if the breathing regime simulated by the flow regulators.

- The two flow regulators are mass flowmeters-controllers.

- The metering pump is located downstream of the second flow regulator on the third fluid line.

Moreover, the present invention also relates to a facility for calibrating a gas analyzer of the type considered here-above, this facility comprising a reservoir of a gas mixture with a predetermined composition, at least one breathing simulator as mentioned here-above, a first pipe connecting an outlet of the gas reservoir to the first inlet of the breathing simulator and a second pipe connecting the outlet of the breathing simulator to an inlet of the gas analyzer.

Advantageously, the gas mixture in the reservoir includes 80% of nitrogen, 16% of oxygen and 4% of carbon dioxide.

According to another advantageous aspect of the invention, the facility also includes an electronic device for computing an oxygen and/or carbon dioxide concentration in a flow of gas exiting the breathing simulator at its outlet, on the basis of some setting parameters of the first and second flow regulators.

According to a third aspect, the present invention relates to a calibration method for calibrating a gas analyzer as the one mentioned here-above, this calibrating method including at least the following steps: a) connecting the gas analyzer to a first inlet of a breathing simulator as mentioned here-above;

b) admitting a quantity of gas mixture from a reservoir into the breathing simulator; c) regulating, via the first flow regulator and while the metering pump is working, the flowrate of gas mixture in the first fluid line of the breathing simulator;

d) regulating, via the second flow regulator and while the metering pump is working, the flowrate of combined gas mixture and air in the third fluid line of the breathing simulator;

e) measuring a first value of the oxygen concentration and/or a first value of the carbon dioxide concentration in the gas flow exiting the breathing simulator, via the gas analyzer;

f) repeating steps c) to e) at different flowrates of gas mixture and/or combined gas mixture and air;

g) computing, for each flowrate of gas mixture and/or each flowrate of combined gas mixture and air, a second value of the oxygen concentration and/or of the carbon dioxide concentration in the gas flow exiting the breathing simulator, based on the flowrates regulated at steps c) and d); and

h) elaborating of a data set making a link, for each flowrate of the gas mixture and/or each flowrate of combined gas mixture and air, between the first measured value measured at step e) and the second computed value computed at step g). Advantageously, this calibration method might incorporate one or several of the following optional features, taken in any admissible combination:

- During steps b) to f), the metering pump works at the same speed, irrespective of the breathing regime simulated by the breathing simulator.

- After step h), the data set is used to correct any concentration measured by the gas analyzer.

- During step 300, the first flow regulator is piloted by the electronic control unit, as a function of a variable flowrate of oxygen consumption, in order to set the flowrate of the gas mixture in the first fluid line of the breathing simulator as at value given by the following equation: Q4 = V0 ₂ / (Fa° ² X F4 ^N2 /Fa ^N2 - F4° ² ) X Q2/Q1 , where Q4 is the flowrate of the gas mixture in the first fluid line of the breathing simulator, V 0 ₂ is a value of the variable rate of oxygen consumption for which the calibration is made, Fa° ² is the oxygen concentration in ambient air, Fa ^N2 is the nitrogen concentration in ambient air, F4° ² is the oxygen concentration the gas mixture of the reservoir, F4 ^N2 is the nitrogen concentration the gas mixture of the reservoir, Q1 is the nominal gas flowrate used for feeding the gas analyzer under normal working conditions and Q2 is the flowrate regulated by the second flow regulator.

According to a fourth aspect, the present invention relates to a determination method for determining a variable rate of oxygen consumption V 0 ₂ and/or of carbon dioxide production V C0 ₂ of a living subject with a gas analyzer calibrated according to the calibration method of the third aspect of the invention, this determination method including at least the following steps:

i) measure of an oxygen concentration and/or a carbon dioxide concentration in gases expired by the living subject, with the calibrated gas analyzer, and j) correction of a measured value obtained at step i) on the basis of the data set.

According to a fifth aspect, the present inventin relates to a diagnostic method for evaluating the energetic dependency of patient, wherein one implements the determination method mentioned her-above and uses the value corrected at step j) for estimating the energetic dependency of the patient.

The invention will be better understood on the basis of the following description of one embodiment of a breathing simulator, a facility, a calibration method and a determination method according to the invention, given only as an example and made on the basis of the following figures:

- figure 1 is a schematic view of a facility according to the invention including a breathing simulator according to the invention; and

- figure 2 is a block diagram of a calibration method according to the invention.

The facility F represented on figure 1 is designed and implemented for calibrating a gas analyzer 6 which includes two cells 6A and 6B respectively provided for measuring the oxygen concentration F° ² and the carbon dioxide concentration F ^c ° ² in a flow of gas provided to this gas analyzer. The cell 6A for measuring 0 ₂ concentration F° ² can make use of paramagnetic technology or chemical analysis, on the basis of zirconium. The cell 6B for measuring C0 ₂ concentration F ^C02 can make use of infrared technology. Alternative techniques can be considered for cells 6A and 6B.

On the basis of the concentrations measured via gas analyser 6, it is possible to compute the rate of oxygen consumption V0 ₂ and the rate of carbon dioxide production V C0 ₂ , as explained here-below. Therefore, gas analyser 6 allows determining V0 ₂ and VC0 ₂ rates indirectly, by computation on the basis of the measured values of concentration F° ² and F ^c ° ² .

Gas analyzer 6 is supposed to be fed with a flow of gas with a non-represented flowrate Q1 equal to 30 liters per minute (l/min). In other words, Q1 is the nominal gas flowrate used for feeding the gas analyzer under normal working conditions, i.e. when it is used after calibration. To this end, gas analyzer 6 includes a non-represented pump capable of generating such a flowrate in a feeding pipe. Actually, flowrate Q1 can vary between 20 and 60 l/min.

A breathing simulator 2 is provided in facility F in order to provide gas analyzer 6 with a gas flow having a given flowrate Q6 corresponding to a fraction of flowrate Q1.

Breathing simulator 2 is provided with a first inlet 22 connected to an outlet 4A of a reservoir 4 which, in this example, is in the form of a steel tank with a capacity of 20 liters under a pressure of about 200 kPa. A gas mixture with a known composition is stored in reservoir 4. The composition of this gas mixture is 80% nitrogen, 16% oxygen and 4% carbon dioxide. In an alternative embodiment of the invention, another composition of the gas mixture stored in reservoir 4 can be considered.

A first pipe 52 connects outlet 4A to inlet 22. Arrow Q4 in pipe 52 represents the flowrate of the known gas mixture transiting from reservoir 4 to breathing simulator 2.

Breathing simulator 2 also includes a second inlet 24, an outlet 26 and a gas mixer

28.

A first fluid line 32 connects first inlet 22 of breathing simulator 2 to a first inlet 282 of gas mixer 28. A second fluid line 34 connects second inlet 24 of breathing simulator 2 to a second inlet 284 of gas mixer 28. A third fluid line 36 connects the outlet 286 of fluid mixer 28 to outlet 26 of breathing simulator 2.

Second inlet 24 opens into the atmosphere so that second fluid line 34 allows feeding gas mixer 28 with ambient air. The flowrate of air in second fluid line 34 is represented by arrow Q3 on figure 1.

The flowrate Q4 of the gas mixture coming from reservoir 4 is the same in first fluid line 32 and in pipe 52.

The gas flow in third fluid line 36 results from the combination of the gas mixture flow in first fluid line 32 and of the air flow in second fluid line 34. Thus, the flowrate Q2 of the gas flow in third fluid line 36 equals the sum of flowrates Q3 and Q4. In other words, the following relationship prevails:

Q2 = Q3 + Q4 (equation 1 )

Gas mixer 28 is represented on figure 1 as a T-joint with a mixing chamber 288 between inlets 282, 284 and outlet 286. Alternatively, other types of gas mixers can be considered.

In order to avoid a back flow of gases, in particular gas mixture coming from reservoir 4, towards the atmosphere, a check valve 44 is mounted on second fluid line 34, between second inlets 24 and 284. Check valve 44 allows gas circulation from inlet 24 to inlet 284, but blocks gas circulation in the reverse direction. A first flow regulator 42 is mounted on first fluid line 32, between first inlets 22 and 282. This first flow regulator 22 is controlled by an electronic control unit or ECU 50 which includes, amongst others, a microprocessor 502 and a memory 504.

Actually, first flow regulator 42 is controlled by ECU 50 via a first electronic signal

S42.

A second flow regulator 46 is mounted on third fluid line 36, between outlets 286 and 26. This second regulator is controlled by ECU 50 via an electronic signal S46.

According to an advantageous aspect of the invention, flow regulators 42 and 46 are of the same kind, that is rely on the same technology. For example, these flow regulators can be EL-FLOW PRESTIGE mass flowmeters-controllers marketed by the company BRONKHORST (France) under reference FG-201 CV-AAD-33-V-AA-. In practice, regulators 42 and 46 are set differently from each other. Regulator 42 is set to control a flowrate between 0 and 0.75 Nl/min (normal liter per minute). Regulator 46 is set to control a flowrate between 0 and 2 Nl/min.

Other apparatuses can be used as flow regulators in breathing simulator 2.

A metering pump 48 is also mounted on third fluid line 36, downstream of second flow regulator 46, that is between flow regulator 46 and outlet 26 of breathing simulator 2. This metering pump 48 is electronically controlled by ECU 50, via an electronic signal S48. Pump 48 is a membrane pump, which has the advantage of being tight and the volume sucked by the pump equals the volume pushed by the pump.

The nominal capacity of metering pump 48 is chosen in order for it to be able to draw, at a given flowrate, a combined gas flow, made of the gas mixture coming from reservoir 4 and from ambient air, from gas mixer 28 toward outlet 26. In practice, this nominal capacity can be chosen equal to about 2.4 l/min of ambient air (ATP: Ambient Temperature Pressure), which corresponds to a standard temperature and pressure working capacity of about 2 l/min, where the standard temperature and pressure working capacity of a pump is its capacity at 0°K, 1013 hPa and 0% hygrometry (STPD: Standard Temperature Pressure Dry). In practice, the standard temperature and pressure working capacity of metering pump 48 can be chosen between 0,5 and 4 l/min, preferably between 1.5 and 3 l/min, the value of 2 l/min being a still preferred value.

A second pipe 54 connects breathing simulator outlet 26 to an inlet 82 of a T-joint 8. A first outlet 84 of T-joint 8 is connected, via a third pipe 56, to a first inlet 62 of gas analyzer 6. A second outlet 86 of T-joint 8 is connected, to the atmosphere, via a fourth pipe 58. The flow of gas transiting through third fluid line 36 and through pipe 54 is divided within T-joint 8 into a first flow entering gas analyzer 6 through pipe 56, with flowrate Q6, and a second flow exiting to the atmosphere through pipe 58, with flowrate Q8. In practice flowrate Q6 is fixed for a given gas analyzer 6, in this example equal to 350 ml/min.

The following relationship prevails:

Q2 = Q6 + Q8 (equation 2)

T-joint 8 may have the same structure as gas mixer 28. However, this is not compulsory.

According to an optional but in practice important aspect of the invention, fluid lines 32, 34 and 36, flow regulators 42 and 46, metering pump 48 and ECU 50 are enclosed within a box 21 , which forms a housing for these parts. This enables carrying and storing breathing simulator 2 in a very convenient way.

Actually, inlets 22 and 24 and outlet 26 are formed in some of the walls of box 21 .

On figure 1 , T-joint 8 is represented outside box 21. In practice, first pipe 54 may be supple and T-joint 8 can be located adjacent one wall of box 21 , on the outside of box 21.

According to an alternative embodiment of the invention, T-joint 8 can be incorporated into box 21 . In such a case, T-joint 8 is connected directly to third fluid line 36 and outlet 26 is combined with inlet 82.

Outlet 4A of tank 4 can be permanently or semi-permanently connected to inlet 22, as long as tank 4 is not empty. A non-represented valve, incorporated in reservoir 4 at the level of outlet 4A, is closed, but when the calibration method of the invention is implemented.

When it is necessary to calibrate gas analyzer 6, its first outlet 62 is connected, in a first step 100 of the calibration method of the invention, to the first outlet 84 of T-joint 8, via pipe 56. At the end of this first step, breathing simulator 2 is connected to gas analyzer 6.

Thereafter, the valve of reservoir 4 is opened and metering pump 48 is piloted by ECU 50 at its nominal capacity. In other words, metering pump 48 is controlled at a given speed, close to its maximum speed, via a constant control tension, so that this pump does not work in a low regime, thus avoiding the risk of generating vibrations within breathing simulator 2.

The respective flowrate Q4 and Q2 in first and second fluid lines 32 and 36 are respectively controlled by flow regulators 42 and 46 which are piloted by ECU 50.

In other words, metering pump 48 works always at the same speed, close to its maximum speed, and flow regulators 42 and 46 are used to lower the respective flowrates Q4 and Q2, as compared to a maximum flowrate obtainable with metering pump 48.

More precisely, second flow regulator 46 is used to adjust the value of flowrate Q2 on a value, set by microprocessor 502, whereas first flow regulator 42 is used to adjust flowrate Q4, on another value set by microprocessor 502, thus the proportion of gas mixture coming out of reservoir 4 into the combined gas mixture and air flow transiting through third fluid line 36.

Indeed, since the pressure drop within second fluid line 34 is constant or substantially constant, the respective proportions of gas mixture coming from tank 4 and air coming from the atmosphere can be regulated by the first and second flow regulators 42 and 46, without adding a flow regulator on second fluid line 34.

However, in a variant, a third flow regulator could be mounted on second fluid Iine34.

Considering a case where gas analyzer 6 is to be calibrated for a session where the patient will be at rest under a canopy, one can reasonably anticipate that the V0 ₂ value to be measured will be between 200 ml/min and 500 ml/min. Thus, calibration of gas analyzer 6 can be implemented for four different values of V 0 ₂ , namely 200 ml/min, 300 ml/min, 400 ml/min and 500 ml/min. Alternatively, the number of values can be different from four, in particular equal to three. Then the values are adapted, for example equal to 200 ml/min, 350 ml/min and 500 ml/min in case of three different values.

When the patient is at rest under the canopy, the flowrate Qext of air exiting the canopy equals the flowrate Qa of air entering the canopy plus the flowrate Qe of gases expired by the patient, minus the flowrate Qi of air inspired by the patient. Thus, the following relationship prevails:

Qext = Qa + Qe - Qi (equation 3)

Fext ^N2 denotes the nitrogen concentration in the air exiting the canopy. Fa ^N2 denotes the nitrogen concentration in the air entering the canopy. Fe ^N2 denotes the nitrogen concentration in the air expired by the patient. Fi ^N2 denotes the nitrogen concentration in the air inspired by the patient. Considering that nitrogen is not consumed by the respiration of the patient, the quantity of nitrogen entering the canopy is the same as the quantity of nitrogen exiting the canopy, which can be expressed as:

Fext ^N2 X Qext = Fa ^N2 X Qa (equation 4)

For the same reason, the quantity of nitrogen inspired by the patient is the same as the quantity of nitrogen expired, which can be expressed as:

Fi ^N2 X Qi = Fe ^N2 X Qe (equation 5)

Equations 4 and 5 can be inverted as follows

Qa = Qext X Fext ^N2 / Fa ^N2 (equation 6)

Qi = Qe X Fe ^N2 / Fi ^N2 (equation 7)

On the other hand, one can assume that the oxygen concentration Fi° ² in the inspired air and the nitrogen concentration Fi ^N2 in the inspired air are respectively the same as the oxygen concentration Fa° ² in ambient air and the nitrogen concentration Fa ^N2 in ambient air. The oxygen consumption rate V0 ₂ equals the oxygen flowrate entering the canopy minus the oxygen flowrate exiting the canopy and can be expressed as follows, when considering the oxygen concentration Fa° ² in the air entering the canopy and the oxygen concentration Fext° ² in the air exiting the canopy:

V0 ₂ = Qa X Fa° ² - Qext X Fext° ² = Qa X Fi° ² - Qext X Fext° ² (equation 8)

In view of equation 6, equation 8 can be rewritten as:

V0 ₂ = Qext X (Fi° ² X Fext ^N2 / Fa ^N2 - Fext° ² ) (equation 9)

According to the same kind of computation, the variable rate of carbon dioxide production VC0 ₂ can be expressed as:

VC0 ₂ = Qext X (Fe ^c ° ² - Fa ^c ° ² X Fext ^N2 / Fa ^N2 ) (equation 10)

Thus, equations 9 and 10 give the V0 ₂ and VC0 ₂ rates as functions of known quantities, namely the volume of extracted air and the measured concentrations in nitrogen, oxygen and carbon dioxide.

For the simulation, tank 4 is used instead of the patient. F4 ^N2 , F4° ² and Q40 respectively denote the nitrogen concentration and the oxygen concentration in the mixture of tank 4 and the flowrate of the mixture coming out of this tank for generating a given V 0 ₂ rate. The same kind of computation as above for V 0 ₂ leads to:

V0 ₂ = Q40 X (Fi° ² X F4 ^N2 /Fi ^N2 - F4° ² ) (equation 1 1 )

This can be also applied for V C0 ₂ as follows:

VC0 ₂ = Q40 X (F4 ^C ° ² - Fa ^c ° ² X F4 ^N2 / Fa ^N2 ) (equation 12)

These two equations can also be inverted in order to express flowrate Q40 as a function of V 0 ₂ and V C0 ₂ , as follows:

Q40 = V0 ₂ / (Fi° ² X F4 ^N2 /Fi ^N2 - F4° ² ) (equation 13)

Q40 = VC0 ₂ / (F4 ^C ° ² - Fa ^c ° ² X F4 ^N2 / Fa ^N2 ) (equation 14)

As mentioned here-above, gas analyzer 6 is supposed to be fed with a gas flow having a flowrate Q1 of, for example, 30 l/min, when it makes some measures, apart from calibration. When it is calibrated, gas analyzer 6 is fed with a flow of gas derived from the gas flow at 30 l/min. The derived gas flow has a flowrate determined by a non-represented pump installed on or near inlet 62. In practice, the flowrate Q6 of the derived flow can be equal to about 350 ml/min.

In such a case, flow regulators 42 and 46 are piloted by ECU 50 via signals S42 and S46 so that flowrate Q2 equals 600 ml/min, that is 50 times less than flowrate Q1 but still more than 350 ml/min. Thus, enough gas enters T-joint 8 to feed gas analyzer 6 for the calibration. In practice, the ratio Q2/Q1 can be chosen between 0,02 and 0,05, preferably equal to 0,02.

Let us consider the case when one needs to simulate a V 0 ₂ of 200 ml. With equation 13, it is possible to compute the partial flowrate Q40 of gas mixture to be delivered by tank 4 for simulating a V 0 ₂ of 200 ml in a global gas flow having a flow rate equal to Q1. Since flowrate Q6 is Q1/Q2 = 50 times less than flowrate Q1 , the partial flowrate Q4 of oxygen in flowrate Q2 can be computed as:

Q4 = Q40 X (Q2/Q1 ) = Q40 / 50

= V0 ₂ / (Fi° ² X F4 ^N2 /Fi ^N2 - F4° ² ) X Q2/Q1

= V0 ₂ / (Fa° ² X F4 ^N2 /Fa ^N2 - F4° ² ) X Q2/Q1 (equation 15)

In addition, the partial flowrate Q3 of ambient air can be computed as

Q3 = Q2-Q40 X (Q2/Q1 ) = Q2-Q40 / 50 (equation 16)

In other words, when a target value of V0 ₂ is known, equations 15 allows microprocessor 502 computing the value of flowrate Q4, depending on the value of V 0 ₂ , Q1 and Q2, thus piloting flow regulators accordingly, via signals S42 and S46. For this computing operation, microprocessor 502 accesses memory 504 in order to collect some data, in particular the concentration values Fi ⁰² , F4 ^N2 , F4 ⁰² .

It should be noted that the value of flowrate Q4 is much smaller than the value of flowrate used in the prior art calibration techniques, which induces substantial savings in the gas mixture of tank 4.

The same approach can be implemented for V C0 ₂ , on the basis of equation 13 and equations similar to equations 14 and 15.

Thus, the method of the invention includes steps subsequent to step 100 and which consist respectively in a step 200 of admission of a quantity of gas mixture from reservoir 4 into breathing simulator 2, a step 300 of regulation of flowrate Q4 via the first flow regulator 42, as explained here-above in connection to equation 15, and another step 400 of regulation of flowrate Q2 via the second flow regulator 46 while metering pump 48 is operating, also as explained here-above.

When steps 200, 300 and 400 are being performed, a flow of gas with a predetermined composition goes through T-joint 8 and is distributed between pipes 56 and 58. The non-represented pump installed at or near inlet 62 guarantees that the gas flow is fed to gas inlet 62 of analyzer 6 at the proper flowrate Q6, namely 350 ml/min in the example.

When breathing simulator 2 is working, which implies that a flow of gas is provided to gas analyzer 6 with flowrate Q6, gas analyzer 6 can measure a first value F° ² m of the oxygen concentration F° ² and a first value F ^C02 m of the carbon dioxide concentration F ^c ° ² in the flow of gaz.

The values F° ² m and F ^C02 m measured by gas analyzer 6 can be stored in a personal computer 12 to which gas analyzer 6 is connected by an electric line 72. S6 denotes an electric signal provided by gas analyzer 6 to computer 12 and including measured values F° ² m and F ^C02 m.

On the other hand, a second electric signal S2, provided by breathing simulator 2, includes some information on the actual composition of the flow of gas provided to gas analyzer 6 with flowrate Q6. In other words, microprocessor 502 can be programmed to compute a second value F° ² c of the oxygen concentration and a second value F ^C02 c of the carbon dioxide concentration in the gas flow exiting breathing simulator 2 at outlet 84, on the basis of some data stored in memory 504 and on the basis of flowrates Q2 and Q4.

This computation of second values F° ² c and FC° ² c is implemented in a further step 600 of the calibration method of the invention which takes place either after or at the same time as step 500.

Steps 200 to 500 can be repeated with different settings of flow regulators 42 and 46, that is with different flowrates Q2 and/or Q4, which corresponds to different simulations of a breathing patient, at rest with a V 0 ₂ ranging between 200 ml/min and 500 ml/min as considered in the example here above, with different proportions of oxygen and carbon dioxide. In other words, steps 200 to 500 can be repeated at different flowrates simulating different types of exhaust gases to be analyzed by gas analyzer 6. Each time, a first value F° ² m of oxygen concentration and a first value F ^C02 m of carbon dioxide concentration is measured via gas analyzer 6 at step 600 and, each time, a second value F° ² c and F ^C02 c is computed at step 600.

In view of the different values measured at the different steps 500 and the different values computed that the different steps 600, it is possible to implement a further step 700 where a data set D is elaborated which makes a link or a connection between each first value F° ² m and/or F ^C02 m and the second value F° ² c and/or F ^C02 c computed for the same settings of flow regulators 42 and 46, that is for the same flowrates Q2 and/or Q4.

This data set D can be used to draw a correspondence table between the measured and computed values and for any other value in the same range, for instance via a linear interpolation or with any other appropriate regression method.

Thanks to breathing simulator 2, a flow of gas, with flowrate Q6 substantially equal to 350 ml/min, can be provided to gas analyzer 6 for the measures of step 500, even if the non-represented pump of gas analyzer 6 normally used for generating flow rate Q1 has a standard temperature and pressure working capacity of 30 l/min or more. Taking the example of a patient at rest, whose expired gases generally have a flowrate between 200 and 500 ml/min, the practitioner can choose to calibrate gas analyzer 6 for successively measuring oxygen concentration F ⁰² at four flowrates, namely 200 ml/min, 300 ml/min, 400 ml/min and 500 ml/min, these values being adjusted via flow regulators 42 and 46, as explained here above. Then, four values of F ⁰² m can be measured at step 500 and four values F ⁰² c can be computed at step 600, forming a data set D with eight values enabling computation of the actual value of F ⁰² detected by gas analyzer 6 on the basis of a linear interpolation among this data set D.

The same approach can be implemented for carbon dioxide concentration F ^c ° ² .

As mentioned here above, the number of flowrates for which measures take place, via gas analyzer, can be different from four, in particular equal to three.

Once gas analyzer 6 has been calibrated with the method represented on figure 2, it is possible to determine the V0 ₂ or VC0 ₂ variable rates of a living subject by measuring the oxygen and/or carbon dioxide concentration with the calibrated gas analyzer, correcting any value measured by this gas analyzer on the basis of the previously elaborated date set D, as explained here-above, and computing the V0 ₂ and/or VC0 ₂ rates on the basis of equations 9 and 10, or similar equations.

According to an alternative embodiment of the invention, instead of computing second value F ⁰² c and/or F ^C02 c in microprocessor 502, data corresponding to the settings of the regulators 42 and 46 can be provided to computer 12 in signal S2 and the actual computation of value can be implemented in this computer.

According to an aspect of the invention which is important in practice, an automatic routine can be used for calibrating gas analyzer 6, this routine including steps 200 to 700, with the repetition of steps 200 to 600 for different flowrates or regimes. These different flowrates or regimes can be preselected by the practitioner or set by the routine, depending on the expected range of flowrates to be provided to the gas analyzer 6 when determining these variable rates after calibration of the gas analyzer.

As mentioned here above, for a patient at rest, the values of 200, 300, 400 and 500 ml/min can be used.