FR2479448A1 - Ultrasonic measurer of plasma ratio in centrifuged blood - uses gating circuit to switch piezoelectric transducer, from transmit to receive mode to determine measurements for subsequent display - Google Patents

Abstract

The heights or ratios of heights of two liquids superimposed in a container are determined by an ultrasonic measuring system. This is used to determine the ratio of plasma in centrifuged blood samples. A test tube is mounted in a supporting sleeve with an acoustic impedance which is approximately equal to that of the test tube. The sleeve is screwed into a base containing a reversible piezoelectric transducer positioned so that the active face is against the base of the test tube. A gelatinous material provides acoustic coupling. A gating circuit allows a pulse generator to be connected to the transducer to emit trains of pulses, and switches into circuit an echo processing circuit to determine the required measurements for display.

Description

METHOD AND DEVICE FOR MEASURING THE REPORT (S)
OF HEIGHTS OF SUPERIMPOSED LIQUIDS AND APPLICATION
FOR THE MEASUREMENT OF THE HEMATOCRITY RATE OF A SPECIMEN
BLOOD MENT
The invention relates to a method and a device for measuring the height ratio (s) of at least two liquids arranged in a container in superimposed layers. It is particularly applicable in the medical field to the measurement of the hematocrit level of a blood sample.

The hematocrit level is the ratio of the height of the globular fluid to the total height of the liquid column (red blood cell and plasma) of a blood sample that has been made incoagulable and centrifuged. The determination of this rate is a very common operation from the medical point of view, since it provides a very interesting indication for a large number of patients, in a state of traumatic shock, medical or surgical.

The measurement method currently most used in practice consists in placing in a capillary tube a portion of the collected blood rendered incoagulable, in performing a centrifugation of this capillary tube and in deducing the hematocrit rate from the height of red blood cell and of the total height of the liquid column thanks to a visual location of the level of the interfaces and a reading of the corresponding rate on an abacus.

 This method repeated at high frequency in specialized laboratories has various drawbacks. In the first place, it requires delicate manipulations of the blood sample, on the one hand to carry out the transfer of a part of it into a capillary tube, on the other hand to execute the reading operations. In addition, it involves risks of errors, to identify the results of a measurement which can be incorrectly attributed to a sample not concerned, or to operate the visual locations and readings subjected to human failures always possible.

The present invention proposes to remedy the drawbacks mentioned above by providing a new method and a new device for performing an automatic measurement of the hematocrit level of a blood sample.

An objective of the invention is in particular to eliminate the operation of transfer into a capillary tube, by authorizing a direct measurement on the original tube which arrives in the laboratory.

Another objective is to eliminate the human operations of level tracking and / reading on charts,
as well as the related manipulations, by providing a signal directly representative of the measurement results, which is capable of being displayed in digital form or of being entered into a computer system.

 More generally, the invention proposes to allow the automatic measurement of one or more height ratios of liquids of different density, contained in superimposed layers in a container, in particular in a common type tube. .

To this end, the measurement method according to the invention consists in sending at least one train of ultrasonic pulses into liquids through one of the extreme interfaces of said liquids, to collect at this extreme interface the ultrasonic pulses reflected by the interfaces. between liquids and by the other extreme interface, to transform these reflected pulses into electrical pulses, to process said electrical pulses to supply signals in number as a function of the number of liquids, first / the nal being representative of the time difference separating the incident train and the ingulaions reflected by the first second interface between liquids, the / signal being representative of the time difference separating the pulses reflected by this first interface and the pulses reflected by the second
of these / me interface ..., to correct each / signals by multiplying it by a coefficient proportional to the speed of propagation of ultrasound in the liquid concerned, and to perform on the signals obtained appropriate calculation operations to generate one or more signals representative of the desired height ratio (s).

The method according to the invention can be applied to carry out an automatic measurement of the hema rate
topped with a blood sample; this sample generally arrives at the laboratory in a glass tube, the blood having been previously / made incoagulable; the centrifugation can be carried out in this original tube without any transfer, and the measurement method implemented in this same tube, which eliminates the manipulations of the conventional methods and the errors of identi
fication.

The train of incident sound pulses
is reflected, on the one hand, on the red blood cell / plasma interface,
on the other hand, on the plasma / air interface. These reflected sound pulses are captured, transformed into electrical pulses
triques and processed to provide two representative signals, one of the time difference between the emission and the first reflection, the other of the time difference between the first and the second reflection. The corresponding signal
to the propagation of ultrasound in the globular liquid is corrected by means of a coefficient proportional to the speed of propagation in this liquid (1680 m / s), while the signal which corresponds to the propagation in the plasma is corrected by means of '' a coefficient proportional to the speed of propagation in the plasma (1500 m / s).

The signals obtained are representative of the heights of globular liquid (L1) and height of plasma
(L2). The hematocrit L1 level is obtained by performing
L1 + L2
on these signals the appropriate calculation operations (summation
and division); the signal representative of this rate can be delivered to a digital display unit and / or to a computer operating system. The sources of errors originating in the conventional methods of the operations of locating
levels and reading on abacuses are thus eliminated.

In addition, according to a mode of setting
preferential work of the process of the invention the treatment
electrical pulses images of the reflected ultrasonic pulses comprises operations of shaping said
electrical pulses, providing substantially rectangular pulses, and an operation of generating,
from these, rectangular window signals
each, of duration equal to the interval separating two consecutive pulses.

 These window signals can in particular be corrected, by means of the coefficients proportional to the propagation velocities in the various liquids, by generating for each signal an oscillating signal of frequency proportional to the corresponding coefficient, by performing a logical AND operation between each window signal and the corresponding oscillating signal in order to obtain an oscillating signal of duration equal to that of the window signal, and by processing the oscillating signals thus obtained to generate signals representative of the number of oscillations thereof.

The invention extends to a measuring device allowing the implementation of the method described; according to the invention, this device comprises
ultrasonic emission means adapted to be applied against the base of a container of liquids in order to periodically deliver trains of ultrasonic pulses to the container / liquids interface, e. an electric generator, able to excite the ultrasonic emission means and associated with a control clock providing the origin of the times for each pulse train,
ultrasonic receiving means adapted to be applied against the base of the container of liquids in order to collect the reflected ultrasonic pulses - and transform them into electrical pulses,
means for processing said pulses suitable for delivering signals representative of the time differences between incident train and reflected pulses,
matching means adapted to deliver signals representative of the product of the signals coming from the processing means with predetermined coefficients,
a calculation unit-receiving the signals from the correction means to carry out logical operations on them making it possible to obtain one or more signals representative of the report or reports sought.

 In a manner known per se, the ultrasonic transmission means and the ultrasonic reception means may comprise a reversible piezoelectric transducer and switching means suitable for connecting, on the one hand, the excitation generator with said transducer in transmission period, on the other hand, said transducer with the processing means during reception period.

 According to another characteristic of the device of the invention, a layer of gelatinous material with an acoustic impedance equal to or close to that of the material constituting the container for liquids is disposed between the piezoelectric transducer and the base of said container in order to provide a connection Continuous acoustics between the transducer and the container and reduce stray reflections at the base of the container.

Furthermore, in the application for measuring the hematocrit level or in similar applications where it is necessary to measure the ratio of the height of a liquid to the total height of two superimposed liquids, the device comprises a summator receiving the signals from the correction means for the purpose of summing them, and a divider receiving one of the signals from these correction means and the signal from the summator for the purpose of dividing the first by the second.

Other characteristics, objects and advantages of the invention will emerge from the description which follows, with reference to the appended drawings, which present, by way of nonlimiting example, an embodiment of the invention, intended in particular for hematrocrit level measurements; on these drawings
FIG. 1 is a block diagram illustrating this embodiment,
FIG. 2 is a timing diagram giving the appearance of the signals coming from the various stages of the device,
FIG. 3 is a schematic detail section of part of the device,
FIG. 4 is a block diagram illustrating in more detail one of the assemblies of the device,
FIG. 5 is a timing diagram giving the appearance of the signals coming from the various units represented in FIG. 4.

 The device symbolically represented in FIG. 1 is intended to measure the rate of hematocrit of blood samples contained in tubes or test tubes such as 1; these test tubes come directly from the sampling center, the blood having been treated, in particular by adding heparin, to make it incoagulable.

 Each specimen is centrifuged in the measurement laboratory, so as to separate the red blood cells and the plasma. The liquid column therefore has, at the time of measurement, a red blood cell / plasma interface and a plasma / air interface.

The hematocrit level is the ratio of the height of the red blood cell to the total height of the liquid column.

To carry out the measurement, the test piece 1 is placed in a support constituted by a sleeve 2 (fig.

3) preferably having an acoustic impedance substantially equal to that of the test piece in order to reduce parasitic echoes.

 This sleeve 2 is threaded at its base and is screwed into a base 3 which contains a reversible piezoelectric transducer 4 connected to an electrical conductor5.

 This transducer 4 is positioned so that its active face (emitting and receiving) is located opposite the bottom of the test piece; this active face is surmounted by a layer of gelatinous material 6 of acoustic impedance equal to or close to that of the test piece. This material provides acoustic conduction between the transducer 4 and the test piece 1 and makes it possible to reduce parasitic echoes.

 The electrical conductor 5 is connected by means of electronic switching means 7, on the one hand to an excitation generator 8, ~ d? ~ On the other hand. means of treatment.

These switching means 7 of conventional type (diode doors or the like) connect either the generator 8 and the transducer 4 when an excitation signal is delivered by said generator, or the transducer 4 and the processing means when no signal is emitted by the generator 8.

 The generator 8 delivers a periodic signal E (Fig. 2) consisting of a brief electric excitation pulse which is reproduced at each measurement cycle; each cycle can last approximately one to two milliseconds. This duration is much greater than the round-trip propagation time of the ultrasonic signal in the liquid column, so that the risks of interference between different cycles are eliminated. A clock 10 defines the origins of each measurement cycle and the interval separating the generation of two excitation pulses.

Each pulse E excites the transducer o which delivers a train of ultrasonic waves to the test piece 1. This train of waves is composed of a series of damped oscillations. These oscillations propagate in the liquid column and undergo partial reflections at the tube bottom / liquid, ear globule / plasma and plasma / air interfaces; parasitic reflections are also produced by the walls of the specimen and by the interfaces after the first round trip of the waves.

 The transducer 4 receives the various echoes and transforms them into an electrical signal R (Fig. 2). This signal is delivered to the processing means which firstly comprise shaping means 9.

 These means 9 comprise a circuit assembly 9a, said sorting assembly, adapted to suppress noise signals on the one hand, by sorting signals of determined frequencies close to the frequency of the acoustic signal (resonance frequency of the transducer 4), d on the other hand by sorting signals of amplitude greater than a threshold.

 This sorting assembly delivers a conformed signal R1 representative of the effective reflections.

 This signal is processed in another set of circuits 9b, called the echo selector. This echo selector is adapted to suppress the pulses corresponding to parasitic reflections, which are located outside of predetermined time ranges.

There are two ranges in the example, one corresponding to the probable instants of arrival of the pulses reflected by the red blood cell / plasma interface, the other to the probable instants of arrival of the pulses reflected by the interface plasma / air. These ranges can be defined by the fact that the test tube contains in practice an amount of liquid always located between two given levels and that the red blood cell / plasma interface is in practice always located between two given levels: the limits of said samples are therefore chosen time ranges the instants corresponding to reflections on these levels so that the parasitic reflections originating in particular from successive reflections or reflections on the walls of the specimen are suppressed. An example of an echo selector capable of fulfill this function.

 There is obtained at the output of the echo selector 9b a signal R2 (FIG. 2) representative of the two significant reflections (red blood cell / plasma interface and plasma / air interface) with their origin established from the emission signal.

The signal R2 is delivered to a window generator 11 comprising two outputs; one of them delivers a window signal S1 of duration equal to the time interval separating the original pulse
and the first reflection pulse, while the other output delivers a window signal Sz of duration equal to the time difference between the first and the second reflection pulse.

This window generator 11 is produced by logic of the conventional type. Note that the echo selector 9b and the window generator 11 can be-in practice nested circuits, the signal R2 not being able to be physically produced; the result S1 and S2 at the outputs of the window generator is then generated by taking into account a succession of information supplied by the echo selector, which is equivalent to that contained in the signal
R2.

The signals S1 and S2 are delivered to AND gates, 12 and 13 with a view to undergoing a correction, taking account of the difference in the speeds of propagation in the globular liquid and in the plasma.

 Each gate 12 or 13 is associated with an electric oscillator 14 or 15 which delivers an oscillating signal of frequency proportional to the speed of propagation in the corresponding liquid; oscillator 15 has a frequency of 1.68 megahertz (speed of propagation in globular liquid: 1680 m / s) and oscillator 14 a frequency of 1.5 megahertz (speed of propagation in plasma 1500 m / s) .

Signals L1 and L2 are thus obtained, the number of oscillations of which is directly representative of the respective heights of the liquids.

The signal L2 is sent directly to a counting unit 16 which delivers a logic signal representative of the height of globular liquid.

 The signals L1 and L2 are also summed in an adder 17 to deliver a signal L1 + L2 whose number of oscillations is representative of the total height of the liquid column; this signal L1 + L2 is in turn processed in a counting unit 18 which delivers a logic signal representative of this total height.

The above-described correction by means of oscillators and counting processing in practice increase the simplicity and reliability of the device, in addition to the interfaces which would have been necessary if these functions had been performed by numerical calculations.

To further simplify the circuits, the processing of the logic signals from the counting units 16 and 18 is preferably carried out after conversion to analog voltages by Digital / Analog converters 19 and 20.

 The voltages from these converters, the amplitudes of which, once stabilized, are proportional to the height of the globular liquid and to the total height of the liquid column, are supplied to an analog divider 21 which supplies a voltage, the amplitude of which, once stabilized, is representative of the hematocrit level.

An analog gate 22 associated with a delay circuit 23 receiving the clock signals eliminates during each measurement cycle the transient period of stabilization of the significant voltage.

 The analog signal from this gate 22 is a constant voltage, the amplitude of which is proportional to the hematocrit rate. This signal can be used either in a display unit 24 after conventional adaptation in an adapter 25 or in a calculation unit or in a computer system.

 FIG. 4 is a detailed block diagram of an embodiment of the shaping means 9, making it possible to benefit from both excellent reliability and great simplicity of structure.

The sorting assembly 9a essentially comprises, arranged in series
a selective amplifier 26 adapted to selectively amplify the signals of frequency close to the natural frequency of the ultrasonic transmission means and reception means,
an integrating threshold detector 27 adapted to suppress signals of value less than a determined threshold and integrate the signals of value greater than this threshold,
a first single-table circuit 28 adapted to generate from the edges of the previous signal shaped pulses,
a low-pass filter 29 adapted to suppress the pulses of duration lower than a determined threshold,
and a monostable circuit 30 adapted to generate from the edges of the previous signal consistent pulses of duration of the order of the duration of the trains of ultrasonic pulses emitted.

 The timing diagram of FIG. 5 presents the appearance of the intermediate signals ra, rb, rc, rd, originating from these various circuits.

 The selective amplifier 26 improves the useful signal / noise ratio; by noise is meant all signals of frequency different from the resonant frequency of the piezoelectric transducer in emission state.

At the reception level, the transducer has insufficient selectivity to oppose further noise, the reflection on the globular liquid / plas ma interface is of low level and merges with the random noise signals, only its frequency allowing discrimination . The selective amplifier 26 achieves this discrimination and its presence is essential to allow the treatment to be properly ensured.

 The integrating threshold detector 27 supplies, from the signal ra, an aligned signal rb which is adapted to be processed by logic circuits.

 The first single-circuit table 28 is a logic circuit of the conventional type which supplies pulses of constant amplitude re, the number and duration of which depend on the signal rob. This signal re lends itself to filtering in view of the signal rb. This signal re reduce the number of parasitic pulses and achieve a single consistent pulse for each reflection on an interface.

This filtering is carried out in the low-pass filter 29 which eliminates the pulses of very short duration, due to erratic tilting of the monostable circuit 28.

This filter 29 delivers an rd signal and can in particular be constituted in a conventional manner by a NAND logic gate whose output is charged by a capacitor.

 The signal rd obtained is processed by the monostable circuit 30, the metastable state of which has a duration substantially equal to that of the trains of ultrasonic pulses emitted. We therefore obtain for each effective reflection a rectangular pulse, of constant duration and amplitude.

The signal R1 from the monostable circuit 30 is delivered to the echo selector 9b which comprises in particular the following assemblies (Fig. 4)
a logic gate 31, receiving the signals from the two monostable circuits of the sorting assembly and adapted to combine these signals and deliver pulses having origins coinciding with those of the pulses of the first monostable circuit,
another monostable circuit 32 receiving the clock signals and the signal from the aforementioned logic gate to deliver a single pulse at the origin of each train of ultrasonic pulses,
a logic circuit 33 receiving the signal
from the logic gate and the signal from the above-mentioned monostable circuit and suitable for delivering pulses of identical duration, of origins coinciding with those of the signals received,
a counting unit 34 receiving the pulses from the aforementioned logic circuit and adapted to count the number of pulses received between each train of ultrasonic pulses and deliver a signal representative of the first pulses received in programmed number corresponding to the number of interfaces.

The timing diagram of FIG. 5 presents the appearance of the intermediate signals re, rf, r g coming from the various circuits of the echo selector 9b.

The pulses of the signal R1 from the monostable circuit 30 have ill-defined origins because the dice
trips occur on the inclined edges of the rd signal, edges whose slopes are not constant.

Logic gate 31 receives from the
circuit 28 the origins not affected by the filtering and, by a combination of the signals rc and R1, delivers pulses reset in origin (the signal R1 plays the role of mask tem
porel and allows to keep only the effective reflections coming from interfaces).

The monostable circuit 32 is intended to perfect the work of the logic gate 31 for the pulses if
originally killed in the cycle. Indeed, at the origin of each cycle
where superimposed emission and reception with a parasitic reflection on the bottom of the tube, we observe a large number of im
urges and the treatments described above can leave
pass parasitic pulses.

The signal r f- from monostable 32 allows
to the logic circuit 33 to perform a filtering at the origin of the
cycle; this circuit 33 delivers a signal rg composed of pulses
whose relative offsets are representative of the actual time offsets between the reflections; moreover, each
signal pulse rg represents effective reflection and
each reflection is represented by a single impulse. The
logic circuit 33 can in particular be constituted by a com
pairing of NAND doors.

 The counting unit 34 is programmed to count down the appropriate number of pulses, corresponding to the origin of the cycle, to the first reflection on the globular liquid / plasma interface and to the first reflection on the plasma / air interface. The following pulses from successive reflections can thus be eliminated.

 The signal R2 already mentioned is representative of the information coming from the counting unit 34.

 The counting unit comprises in particular a programmed counter and means for erasing the non-counted pulses, means produced in a known manner by a monostable circuit and a combination of logic gates, which make it possible to lock the counting until the next cycle.

Of course, the invention is not limited to the terms of the preceding description but includes all of its variants.

Claims (13)

 1 / - Method for measuring the height ratio (s) of at least two liquids of different density, superimposed in a container, characterized in that it consists in sending in the liquids at least one train of ultrasonic pulses through a extreme interfaces of said liquids, to capture at this extreme interface the ultrasonic pulses reflected by the interfaces between liquids and by the other extreme interface, to transform these reflected pulses into electrical pulses, to process said electrical pulses to provide signals in number function of the number of liquids, the first signal being representative of the time difference separating the incident train and the pulses reflected by the first interface between liquids, the second signal being representative of the time difference separating the pulses reflected by this first interface and the pulses reflected by the second interface, ..., to correct each of said signals by multiplying by a coefficient proportional to the speed of propagation of ultrasound in the liquid concerned, and performing on the signals obtained the appropriate calculation operations to generate one or more signals representative of the desired height ratio (s).  2 / - Measuring method according to claim 1, characterized in that the processing of the electrical pulses images of-s reflected ultrasonic pulses comprises operations operations conformation of said electrical pulses, providing substantially rectangular pulses, and an operation consisting in generating, from these latter, rectangular window signals each of duration equal to the interval separating two consecutive pulses, said window signals being corrected by generating for each signal an oscillating signal of frequency proportional to the speed of propagation in the liquid concerned, by performing an AND logical operation between each window signal and the corresponding oscillating signal in order to obtain an oscillating signal of duration equal to that of the window signal and by processing said osil ltrtt signals thus obtained to generate signals representative of the number of oscillations of these  3 / - Measuring method according to claim 2, c3ractérisé in that the processing of electrical impulses mRge. reflected ultrasonic pulses includes  a sorting operation making it possible to suppress the noise signals and to amplify the signals representing effective flexions, an echo selection operation makes it possible both to suppress the reflected parasitic pulses, located outside of several time ranges, each range corresponding to the probable instants of arrival of the pulses reflected on an interface. 4 / - Application of the method according to one of claims', 2 or 3, for measuring the hematocrit level of a blood sample, centrifuged to be in the form of a column of globular liquid surmounted by a plasma column, the correction of the above-mentioned signals two in number corresponding to the propagation of ultrasound in the globular liquid and in the plasma being carried out by means of coefficients proportional respectively to the speed of propagation of the ultrasound in the globular liquid (1680 m / s) and at the speed of propagation of ultrasound in the plasma (1500 m / s)  5 / - Device for measuring one or more height ratios of at least two superimposed liquids, by implementing the method according to one of claims 1, 2 or 3, said device in particular applicable for measuring the rate hematocrit being characterized in that it comprises  ultrasonic emission means adapted to be applied against the base of a container (1) of liquids in order to periodically deliver trains of ultrasonic pulses to the container / liquids interface,  an electric generator (8), able to excite the ultrasonic emission means and associated with a control clock (10) providing the origin of the times for each train of pulses,  ultrasonic receiving medium waves adapted to be applied against the base of the container of liquids in order to capture the reflected ultrasonic pulses and transform them into electrical pulses,  processing means (9, 11) of said pulses adapted to deliver signals representative of the time differences between incident train and reflected pulses,  correction means (12, 13, 14, 15, 16, 18) suitable for delivering signals representative of the product of the signals coming from the processing means with predetermined coefficients,  a calculation unit (17, 19, 20, 21) concealing the signals coming from the correction means for carrying out on them logical operations making it possible to obtain one or more signals representative of the sought ratio (s).  6 / - Measuring device according to claim 5, wherein the ultrasonic emission means and the ultrasonic reception means comprise a reversible piezoelectric transducer (4) and switching means (7) adapted to connect, on the one hand, the excitation generator (8) with said transducer during the transmission period, on the other hand, said transducer with the processing means during the reception period.  7 / - Measuring device according to claim 6, wherein the liquids are arranged in a tube, characterized in that it comprises a tube support (2, 3) provided with a substantially vertical housing, of suitable shape to contain said tube, the piezoelectric transducer (4) being positioned at the base of said housing facing it and surmounted by a layer of aatinous material (6) with an acoustic impedance equal to or close to that of the tube in view of 'ensuring the acoustic connection between said transducer and said tube.  8 / - Measuring device according to one of claims 5, 6 or 7, characterized in that the processing means comprise  shaping means (9) suitable for shaping the electrical pulses from the ultrasonic reception means and for receiving the origins of time supplied by the clock in order to deliver shaped pulses synchronous with the origins of time and the pulses from the aforementioned means of reception,  a window generator (11) comprising outputs in number as a function of the number of liquids superimposed in the container and adapted to deliver at each output a window signal, of duration equal to the time difference separating two consecutive shaped pulses,  a window signal correction unit (12, 14 or 13, 15) being associated with each output of the window generator and adapted to deliver a signal representative of the product of this window signal with a predetermined coefficient. 9 / - Measuring device according to claim 8, characterized in that each correction unit associated with an output of the window generator comprises an electric oscillator (14 or 15) of frequency proportional to the above-mentioned coefficient, and a logic gate AND (12 or 13) receiving the signal from said oscillator and the window signal to deliver an oscillating signal at the above-mentioned frequency of duration equal to that of said window signal, counting units 16, 18) being provided to receive the signals thus obtained and deliver signals representative of the number of oscillations thereof. 10 / - Measuring device according to one of claims 8 or 9, characterized in that the shaping means comprise a sorting assembly (9a) receiving the electrical pulses from the ultrasonic reception means and adapted to suppress noise signals and delivering pulses representative of effective reflections, and an echo selector (9b) receiving the latter pulses and adapted to suppress the pulses corresponding to parasitic reflections, 11 / - Measuring device as sold cation 10, characterized in that the aforementioned sorting assembly includes, in series:  a selective amplifier (26) adapted for  selectively amplify signals of frequency close to the natural frequency of the transmission means and reception means  ultrasonic,  an integrating threshold detector (27)  suitable for suppressing signals of value less than one determined threshold and integrate higher value signals  at this threshold, a first monostable circuit (28) adapted to generate from the fronts of the previous signal consistent pulses, a low-pass filter (29) suitable for sup priming pulses of duration below a deter threshold  mine,  and a second monostable circuit (30) adap  tee to generate from the fronts of the previous signal consistent pulses of duration of the order of the duration of  trains of ultraspnore pulses emitted.  12 / - Measuring device as sold cation 11, characterized in that the echo selector comprises a logic gate (31), receiving the si  of the two monostable circuits in the set of sorting and adapted to combine these signals and deliver pulses having origins coinciding with those of the pulses of the first monostable circuit, another monostable circuit (32) receiving  clock signals and the signal from the logic gate  above to deliver a single impulse to the origin of  each train of ultrasonic pulses, 5. a logic circuit (33) receiving the signal  from the logic gate and the signal from the monos circuit  aforementioned table and suitable for delivering pulses of duration identical, of origins coinciding with those of the signals received, a counting unit (34) receiving the i.mpulsionr. bones of the aforementioned logic circuit and adapted to count 1 number of pulses received between each train of ultrasonic pulses and deliver a signal representative of the first pulses received in programmed number corresponding to the number of interfaces  13 / - Measuring device according to one of claims 5, 6, 7, 8, 9, 10, it or 12, for measuring in a container containing two superimposed liquids, the ratio of the height of one of them at the total height of the two liquids, said device being characterized in that it comprises a summator (17) receiving the signals from the correction means with a view to effecting the sum thereof, tt tt a divider (21) receiving a signals from the correction means and the signal from the adder in order to carry out the distribution of the first by the second.