1. A system (1) for the in vitro detection andlor quantification by fluorometry of at least one analyte in a sample (E) of fluid, comprising: - a radiation source (10) emitting a main beam (FP) in a given wavelength called emission wavelength; - an optical splitter (13) arranged at the output of the radiation source (10) for splitting the main beam (FP) into a first sample (E)-energizing beam (FE) and a second reference beam (FR); - a first photodetector (14) means designed for providing a first analog detection signal (SADI) in response to detecting a fluorescence ray (RF) emitted by the sample (E), in a so-called fluorescence wavelength as a result of the excitation induced by the first energizing beam (FE); - a second photodetector (15) means designed for providing at the output a 15 second analog detection signal (SAD2) in response to a detection of the second reference beam (FR); said system (1) being characterized in that it comprises: - a generator (300) outputting a sinusoidal carrier signal (SNM) at a predefined frequency called carrier frequency (f,), and at least one digital 20 demodulation signal (SINE, COSINE) at this same carrier frequency (f,); - a digitallanalog conversion means (303) connected to the generator (300) for converting the sinusoidal carrier signal (SNM) into an analog modulation signal (SAM) at the carrier frequency (fo); - an amplitude modulator (4) connected to the digitallanalog conversion 25 means (303) and to the radiation source (10) to modulate in amplitude the main beam (FP) at the carrier frequency (fo) by applying the analog modulation signal (SAM) on said radiation source (10); - analogldigital conversion means (21, 22) connected to the photodetector means (14, 15) to convert the first analog detection signal (SADI) into a first digital 30 so-called fluorescence signal (SFN) and the second analog detection signal (SAD2) into a second digital reference signal (SRN); - digital processing means (32, 33) connected to the generator (300) and to the analogldigital conversion means (21, 22), designed, on the one hand to process the first digital fluorescence signal (SFN) by demodulation at the carrier frequency (fo) 3s in order to calculate a first so-called fluorescence value (VALF) characteristic of the amplitude of the fluorescence ray (RF) and, on the other hand, process the second digital reference signal (SRN) by demodulation at the carrier frequency (fo) in order to calculate a second so-called reference value (VALR) characteristic of the amplitude of the reference beam (FR); - a means for comparing the first fluorescence value (VALF) and the second reference value (VALR) to calculate a final result (RFU) for establishing the detection 5 andlor quantification of the analyte.

2. The system (1) according to claim 1, wherein the digital processing means (32, 33) comprise: - a first demodulation means (34) designed for demodulating the first digital fluorescence signal (SFN) by multiplying it by at least one digital demodulation signal 10 (SINE, COSINE) at the carrier frequency (fo), in order to generate at least one first demodulated fluorescence signal (SFSIN,S FCOS)a;n d - a first calculation (353) means designed to calculate, based on at least one first demodulated fluorescence signal (SFSIN, SFCOS), one first so-called fluorescence value (VALF) characteristic of the amplitude of the fluorescence ray 15 (RF);

3. The system (1) according to claim 2, wherein the first demodulation means (34) comprises: - a first multiplier (341) by a digital demodulation signal (SINE) at the carrier frequency (fo) and in phase with the sinusoidal carrier signal (SNM), said first 20 multiplier (341) being possibly followed by a first low-pass filter (351) at a cut-off frequency (f,) lower than the carrier frequency (fo), in order to output a first so-called in phase demodulated fluorescence signal (SFSIN); and - a second multiplier (342) by a digital demodulation signal (COSINE) at the carrier frequency (fo) and in phase quadrature with respect to the sinusoidal carrier 25 signal (SNM), said second multiplier (342) being possibly followed by a second lowpass filter (352) at the same cut-off frequency (f,), in order to output a first so-called phase quadrature demodulated fluorescence signal (SFcos); and wherein the first calculation means (353), at the output of said first and second multipliers (341, 342), calculates the first fluorescence value (VALF) 30 corresponding to the module of the sum of the first in phase demodulated fluorescence signal (SFSIN) and the first phase quadrature demodulated fluorescence signal (SFcos).

4. The system (1) according to claim 3, wherein the digital processing means (32, 33) comprise, at the input of the first and second multipliers (341, 342): - a notch filter (361) centered on the carrier frequency (fo) in order to filter the first digital fluorescence signal (SFN) and output a first filtered intermediate signal (SFN'); and - a subtracter (362) performing the subtraction of the first digital fluorescence signal (SFN) and the first filtered intermediate signal (SFN'), in order to generate a first filtered fluorescence signal (SFNO), this first filtered fluorescence signal (SFNO) being injected in said first and second multipliers (341, 342) of the first demodulation means (34).

5. The system (1) according to any one of the preceding claims, wherein the digital processing means (32, 33) comprise: - a second demodulation means (37) designed to demodulate the second digital reference signal (SRN), by multiplying it by at least one digital demodulation signal (SINE, COSINE) at the carrier frequency (fo), in order to generate at least one second demodulated reference signal (SRSIN,S RCOS)a; nd - a second calculation means (383) designed to calculate, based on at least one second demodulated reference signal (SRSINI SRCOS)t,h e second reference value (VALR).

6. The system (1) according to claim 5 wherein the second demodulation means (37) comprises: - a first multiplier (371) by a digital demodulation signal (SINE) at the carrier frequency (fo) and in phase with the sinusoidal carrier signal (SNM), said first multiplier (371) being possibly followed by a first low-pass filter (381) at a cutoff 20 frequency (f,) lower than the carrier frequency (fo), so as to output a second so-called in phase demodulated reference signal (SRSIN); and - a second multiplier (372) by a digital demodulation signal (COSINE) at the carrier frequency (fo) and in phase quadrature with respect to the sinusoidal carrier signal (SNM), said second multiplier (372) being possibly followed by a second low- 25 pass filter (382) at the same cutoff frequency (f,) so as to output a second so-called phase quadrature demodulated reference signal (SRcos); and wherein the second calculation means (383), at the output of said first and second multipliers (371, 372), calculates the second reference value (VALR) corresponding to the module of the sum of the second in phase demodulated 30 reference signal (SRSIN) and the second in phase quadrature demodulated reference signal (SRcos).

7. The system (1) according to claim 6 wherein the digital processing means (32, 33) comprise, at the output of the second analogldigital conversion means (22) and at the input of said first and second multipliers (371, 372): 3 5 - a notch filter (391) centered on the carrier frequency (fo) so as to filter the second digital reference signal (SRN) and output a second filtered intermediate signal (SRN'); and - a subtracter (392) performing the subtraction of the second digital reference signal (SRN) and the second filtered intermediate signal (SRN') so as to generate a second filtered reference signal (SRNo), this second filtered reference signal (SRNo) being injected in said first and second multipliers (371, 372) of the second demodulation means (37).

8. The system (1) according to any one of the preceding claims, further comprising, interposed between the radiation source (1 0) and the optical splitter (I3) , an optical band-pass filter (1 1) centered substantially on the emission wavelength.

9. The system (1) according to any one of the preceding claims, further comprising, interposed between the sample (E) and the first photodetector means (14), an optical band-pass filter (141) centered substantially on the fluorescence wavelength.

10. The system (1) according to claim 9, further comprising, interposed between the optical band-pass filter (141) and the first photodetector means (14), a waveguide (142) achieved in the form of a guiding cone.

11. The system (1) according to any one of the preceding claims, further comprising, interposed between the optical splitter (1 3) and the second photodetector means (15), an optical low-pass filter (151) which exhibits a low-wavelength cutoff substantially lower than the emission wavelength.

12. The system (1) according to claim 11, further comprising, interposed between the optical low-pass filter (151) and the second photodetector means (15), a waveguide (1 52) achieved in the form of a guiding cone.

13. The system (1) according to any one of the preceding claims, wherein the sinusoidal carrier signal (SNM) is in the form of a set of several periodic sinusoidal iterations at the carrier frequency (fo), the time difference between two consecutive iterations being higher than the period of sinusoidal iterations.

SYSTEM FOR IN VITRO DETECTION AND/OR QUANTIFICATION BY
FLUOROMETRY
The present invention relates to a system for the in vitro detection and/or
5 quantification by fluorometry of at least one analyte in a sample (E) of fluid
constituting a biological material.
The present invention can be used in automated instruments for in vitro
diagnosis in the clinical or the industrial field.
In the clinical field, the diagnosis is carried out from a human biological sample
10 (urine, blood, saliva, pus, cerebrospinal fluid, etc.), for detecting or quantifying
analytes from an external micro-organism or not (bacteria, virus, parasite, antibody,
etc.) in the form of immunological tests or molecular biology tests.
In the industrial field, the diagnosis is carried out from a sample of food,
pharmaceutical or cosmetic product to control the microbiological quality of the
15 product in the form of microbiological tests. Such microbiological tests generally
check either the sterility (no microorganism should be present), or the absence of
pathogenic bacteria (source of an infection), or even that a commensal bacteria
(normally present in humans and banal in low concentration) is only present below a
certain threshold.
20 The invention can also be used in the field of dynamic analyses, that is to say
in real time, in this instance in immunological tests and molecular biology tests where
the immunological/biological reactions may be managed or controlled over time.
The invention more particularly relates to a system for the in vitro detection
and/or quantification by fluorometry comprising:
2 5 - light radiation source emitting a main beam in a given wavelength called
emission;
- an optical splitter arranged at the output of the radiation source for splitting
the main beam into a first sample-energizing beam and a second reference beam;
- a first photodetector means designed for providing a first analog detection
30 signal in response to detecting a fluorescence ray emitted by the sample, in a socalled
fluorescence wavelength as a result of the excitation induced by the first
energizing beam;
- a second photodetector means designed for providing at the output a
second analog detection signal in response to a detection of the second reference
35 beam.
Such in vitro detection and/or quantification systems, are particularly known in
automated in vitro diagnosis instruments by fluorometry, such as for example in
documents EP 0864 089 B1, EP 0871 863 B1, EP 0 241 268 Al, and WO
20041055502 A2, which use radiation sources of pulsed light source type, laser lamps
or arc lamps.
Typically, the analysis of signals from the two photodetector means is carried
out by an analog processing, with, in particular an algorithm for detecting the
5 fluorescence peak representing the presencelquantity of analytes to detect. The
drawback of such an analog processing substantially resides in its limitations for
reducing noise and hence in increasing the signal-to-noise ratio.
The prior art may also be illustrated by the teaching of patent application US
5 757 013 A which discloses an apparatus for measuring fluorescence decay, which
10 implements a digital processing based on a fluorescence signal and a reference
signal, where the reference signal substantially serves to measure a phase shift for
the fluorescence signal, with the implementation of the calculation of an internal
product between the fluorescence and reference signals demodulated at a frequency
different from the carrier frequency of the drive signal of the light source; this
15 calculation allowing to determine such a phase shift and to deduce the fluorescence
decay thereof. The purpose of the measuring apparatus of application US 5 757 013
A is to determine the fluorescence decay, without reference to the concentration of
fluorophore in the sample.
The prior art also includes the teaching of document WO 00137850 A1 which
20 relates to an apparatus intended to measure the emission delay time during the
irradiation of a sample, comprising a system generating first and second digital input
signals, a device for converting these two signals into analog sinusoidal signals, a
radiation source modulated at a specific frequency and which irradiates the sample,
thus generating an emission by the sample. The apparatus also comprises a detector
25 of the emission from the sample and which generates a first output signal having a
phase shift with respect to the phase of the first input signal, and a device which
allows to digitize the first and second analog output signals, a mixer receiving the
output signals and comparing the signal phase in order to generate a signal
indicating the phase change. A feedback device distorts, on the basis of the mixer
30 signal, the phase of the second input signal in order to place the first and second
output signals in quadrature.
The purpose of the present invention is to provide an in vitro detection andlor
quantification system by fluorometry comprising signal processing means ensuring a
detectionlquantification of the analyte with an improved sensitivity obtained by an
35 increase in the signal-to-noise ratio.
In fact, it proposes an in vitro detection andlor quantification system by
fluorometry of at least one analyte in a sample of fluid, including:
- a radiation source emitting a main beam in a given wavelength called
emission;
- an optical splitter arranged at the output of the radiation source for splitting
the main beam into a first sample-energizing beam and a second reference beam;
- a first photodetector means designed for providing a first analog detection
signal in response to detecting a fluorescence ray emitted by the sample, in a socalled
fluorescence wavelength as a result of the excitation induced by the first
energizing beam;
- a second photodetector means designed for providing at the output a
10 second analog detection signal in response to a detection of the second reference
beam.
This system being remarkable in that it includes:
- a generator outputting a sinusoidal carrier signal at a so-called predefined
carrier frequency, and at least one digital demodulation signal at this same carrier
1.5 frequency;
- a digitallanalog conversion means connected to the generator for
converting the sinusoidal carrier signal to an analog modulation signal at the carrier
frequency;
- an amplitude modulator connected to the digitallanalog conversion means
20 and to the radiation source to modulate in amplitude the main beam at the carrier
frequency by applying the analog modulation signal on said radiation source;
- digitallanalog conversion means connected to the photodetector means to
convert the first analog detection signal into a first digital so-called fluorescence
signal and the second analog detection signal into a second digital reference signal;
2 5 - digital processing means connected to the generator and to the
analogldigital conversion means, designed, on the one hand to process the first
digital fluorescence signal by demodulation at the carrier frequency in order to
calculate a first so-called fluorescence value characteristic of the amplitude of the
fluorescence ray and, on the other hand, process the second digital reference signal
30 by demodulation at the carrier frequency in order to calculate a second so-called
reference value characteristic of the amplitude of the reference beam;
- a means for comparing the first fluorescence value and the second
reference value to calculate a final result for establishing the detection andlor
quantification of the analyte.
Thus, the system according to the invention performs a digital processing by
amplitude modulation and demodulation, allowing to reach particularly interesting
signal-to-noise ratios.
Such a digital processing is preferably carried out with a light-emitting diode
(LED), in particular, in the UV spectrum, as radiation source, as light-emitting diodes
are particularly suitable for use with a modulated drive. With a light-emitting diode, it
may be particularly ensured a set of light flashes of amplitude of well defined and
controlled temporal and frequential forms, the flashes capable of being spaced
temporally apart from each other in a controlled and precise manner. With such a
drive possibility of the light-emitting diode, the following digital processing allows to
rejectllimit noise.
It is understood that the first digital fluorescence signal and the second digital
reference signal are demodulated separately at the same frequency corresponding to
the carrier frequency of the sinusoidal carrier signal prior to carrying out the
comparison which will allow to extract the final result which preferably comes in the
form of a relative fluorescence unit (RFU).
In a particular embodiment, the digital processing means comprise:
- a first demodulation means designed for demodulating the first digital
fluorescence signal by multiplying it by at least one digital demodulation signal at the
carrier frequency, in order to generate at least one first demodulated fluorescence
signal; and
- a first calculation means designed to calculate, based on at least one first
demodulated fluorescence signal, the first fluorescence value.
According to a feature, the fist demodulation means comprises:
- a first multiplier by a digital demodulation signal at the carrier frequency
and in phase with the sinusoidal carrier signal, said first multiplier being possibly
followed by a first low-pass filter at a cut-off frequency lower than the carrier
frequency, in order to output a first so-called in-phase demodulated fluorescence
signal; and
- a second multiplier by a digital demodulation signal at the carrier
frequency and in phase quadrature with respect to the sinusoidal carrier signal, said
second multiplier possibly being followed by a second low-pass filter at the same cutoff
frequency, in order to output a first so-called phase quadrature demodulated
fluorescence signal;
and furthermore, the first calculation means, at the output of said first and
second multipliers, calculates the first fluorescence value corresponding to the
module of sum of the first in phase demodulated fluorescence signal and the first
phase quadrature demodulated fluorescence signal.
With this feature, the demodulation process corresponds to a coherent
demodulation in phase and in phase quadrature which is particularly advantageous
for improving the signal-to-noise ratio.
The system possibly includes at least one low-pass filter, at a cut-off frequency
lower than the carrier frequency, at the output of the first demodulation means in
order to filter the or each first demodulated fluorescence signal.
According to another feature, the digital processing means comprise, in input
5 of the first and second multipliers:
- a notch filter centered on the carrier frequency in order to filter the first
digital fluorescence signal and output a first filtered intermediate signal; and
- a subtracter performing the subtraction of the first digital fluorescence
signal and the first filtered intermediate signal, in order to generate a first filtered
10 fluorescence signal, this first filtered fluorescence signal being injected in said first
and second multipliers of the first demodulation means.
The notch filter associated with the subtracter forms in the end a band-pass
filter excluding the frequential incoming component at the carrier frequency.
In a particular embodiment, the digital processing means include:
- a second demodulation means designed to demodulate the second digital
reference signal, by multiplying it by at least one digital demodulation signal at the
carrier frequency, in order to generate at least one second demodulated reference
signal; and
- a second calculation means designed to calculate, based on at least one
20 second demodulated reference signal, the second reference value.
In this embodiment, the second processing means applies a demodulation
processing on the second digital reference signal, with the advantages pertaining to
this type of process for improving the signal-to-noise ratio.
According to a possibility of the invention, the second demodulation means
25 includes:
- a first multiplier by a digital demodulation signal at the carrier frequency
and in phase with the sinusoidal carrier signal, said first multiplier being possibly
followed by a first low-pass filter at a cutoff frequency lower than the carrier
frequency, so as to output a second so-called in phase demodulated reference
30 signal; and
- a second multiplier by a digital demodulation signal at the carrier
frequency and in phase quadrature with respect to the sinusoidal carrier signal, said
second multiplier being possibly followed by a second low-pass filter at the same
cutoff frequency so as to output a second so-called phase quadrature demodulated
35 reference signal;
and the second calculation means, at the output of said first and second
multipliers, calculates the second reference value corresponding to the module of the
sum of the second in phase demodulated reference signal and the second in phase
quadrature demodulated reference signal.
According to another possibility of the invention, the digital processing means
comprise, in input of these first and second multipliers:
- a notch filter centered on the carrier frequency so as to filter the second
digital reference signal and output a second filtered intermediate signal;
and
- a subtracter performing the subtraction of the second digital reference
signal and the second filtered intermediate signal so as to generate a second filtered
10 reference signal, this second filtered reference signal being injected in said first and
second multipliers of the second demodulation means.
Advantageously, the system further comprises, interposed between the
radiation source and the optical splitter, an optical band-pass filter substantially
centered on the emission wavelength.
The use of such an optical band-pass filter allows to increase the signal at the
photodetector means, by improving at the source the signal by filtering the interfering
frequencies.
According to a feature, the system further comprises, interposed between the
sample and the first photodetector means, an optical band-pass filter substantially
20 centered on the fluorescence wavelength.
The use of such an optical band-pass filter allows to increase the signal at the
first photodetector means, by filtering the interfering frequencies.
According to a feature, the system further comprises, interposed between the
optical band-pass filter and the first photodetector means, a waveguide achieved in
25 the form of a guiding cone.
In a particular embodiment, the system further comprises, interposed between
the optical splitter and the second photodetector means, an optical low-pass filter
which exhibits a low-wavelength cutoff substantially lower than the emission
wavelength.
30 According to a possibility of the invention, the system further comprises,
interposed between the optical low-pass filter and the second photodetector means,
a waveguide achieved in the form of a guiding cone.
According to another possibility of the invention, the sinusoidal carrier signal is
in the form of a set of several periodic sinusoidal iterations at the carrier frequency,
35 the time difference between two consecutive iterations being higher than the period
of sinusoidal iterations.
Other characteristics and advantages of the present invention will become
apparent upon reading the following detailed description, of a non limiting
implementation, made with reference to the accompanying drawings in which:
- fig.1 is a schematic view of a system in accordance with the invention;
- fig.2 is a schematic view of a system in accordance with the invention
according to a first configuration;
- fig.3 is a schematic view of a system in accordance with the invention
according to a second configuration;
- fig.4 is a graph respectively illustrating the transmission spectrum of an
10 optical band-pass filter and the emission spectrum of a UV diode fitting both a system
in accordance with the invention, and the energizing spectrums 4-MU and 4-MUP;
- fig.5 is a schematic graph of a transmission spectrum of an optical splitter
fitting a system in accordance with the invention;
- fig.6 is a schematic graph of the variation of the photosensitivity of a
15 photodiode fitting a system in accordance with the invention, based on the
wavelength of the detected radiation at a temperature of 25°C;
- fig.7 is a schematic graph of the fluorescence spectrum of 4-MU, and the
transmission spectrum of another optical band-pass filter and a guiding cone both
fitting a system in accordance with the invention;
2 o - fig.8 is a schematic view of a system in accordance with the invention
illustrating its signal processing portion, according to a first configuration;
- fig.9 is a graph illustrating the variation curves according to the time of the
sinusoidal carrier signal SNM, of the first digital fluorescence signal SFN and the
second digital reference signal SRN, these signals come in the form of four
25 sinusoidal periodic iterations;
- fig.10 is a graph illustrating an iteration for the three curves of fig.9;
- fig.1 I is a schematic view of two frames transporting the digital data for the
system in accordance with the invention;
- fig.12 is a schematic view of a system in accordance with the invention
30 illustrating its signal processing portion, according to a second configuration
corresponding to an improvement of the first configuration;
- fig.13 is a graph illustrating the variation curve of RFU according to the
concentration in 4-MU in the sample;
- fig.14 is a graph illustrating the variation curve of the drive current ID(t)
35 according to the time t for a pulse;
- fig.15 is a graph illustrating the variation curve of the drive current ID(t)
according to the frequency f;
- fig.16 is a graph illustrating the variation curve of the first analog detection
signal SAD1 (t) according to the time t for a pulse;
- fig.17 is a graph illustrating the variation curve of the first digital
fluorescence signal SFN(f) according to the frequency f;
- fig.18 is a graph illustrating the variation curve of the function HNOTCH(f )
according to the frequency f;
- fig.19 is a graph illustrating the variation curve of the first filtered
fluorescence signal SFNo(f) according to the frequency f;
- fig.20 is a schematic view of a set of two multipliers, two low-pass filters
10 and a calculation means for a first module for acquiring/processing the digital
fluorescence signal;
- fig.21 is a graph illustrating the variation curve of the intermediate
demodulated signal in phase SFSIN~(af)c cording to the frequency f;
- fig.22 is a graph illustrating the variation curve of the intermediate
is demodulated signal in phase quadrature SFcosl(f) according to the frequency f;
- fig.23 is a graph illustrating the variation curve of the function HLpsmooth(f)
according to the frequency f;
- fig.24 is a graph illustrating the variation curve of the first demodulated
signal in phase SFSIN(f) according to the frequency f;
2 o - fig.25 is a graph illustrating the variation curve of the first demodulated
signal in phase quadrature SFcos(f) according to the frequency f;
The first part of the following description relates to the architectural or
structural part of the system 1 in accordance with the invention for the in vitro
detection and/or quantification by fluorometry of at least one analyte in a sample E of
25 fluid constituting a biological material. The figs. 1 to 3 schematically illustrate the
architecture of such a system 1 in accordance with the invention and designed for
fitting an automated in vitro diagnosis instrument.
This system 1 is intended in particular for analyzing by fluorometry the
fluorescent radiation emitted during an immunological test based on 4-
30 methylumbelliferone (4-MU), as resulting from the hydrolysis of the substrate 4-
methylumbelliferyl-phosphate (4-MUP) in the sample E of fluid. Thus, the function of
this system is to perform an instantaneous measurement of the fluorescent radiation
emitted by the 4-MU in the sample El under the incidence of the energizing beam
suitable for exciting the 4-MU, while avoiding to excite the substrate 4-MUP.
This system 1 comprises a rack B supporting a light radiation source 10
emitting a main beam FP. The radiation source 10 is constituted of an
electroluminescent diode emitting in the ultraviolet (UV LED).
The curves C3 and C4 of fig.4 illustrate respectively the energizing spectrums
of 4-MUP and 4-MU, according to the wavelength (in nm), which exhibit excitation
peaks respectively at 350 nm and 365-370 nm. In order to obtain an emission of a
fluorescence ray of 4-MU, it is thus necessary to have a diode which emits a main
5 beam FP in the wavelength range of 365-370 nm.
Although the maximum of the excitation peak of 4-MU is located at 365 nm, it
is nevertheless preferable that the main beam FP be centered around 370 nm. In
fact, a wavelength of 365 nm for the main beam would excite the 4-MUP too much
and thus cause an interference detrimental to the proper analysis of the fluorescence
10 signal of the 4-MU. The electroluminescent diode 10 thus preferably emits in a
wavelength of around 370 nm; a diode being theoretically monochromatic.
For example, the diode 10 is constituted of a reference diode "NSHU591A
Rank 6" commercialized by the NlCHlA Corporation, emitter in the wavelength range
370-375 nm, with a spectral half-width of 15 nm. The curve C2 of fig.4 illustrates the
is spectrum of the reference diode NSHU591A according to the wavelength (in nm).
In order to answer the issue of limiting the main beam FP wavelength around
370 nm, the system 1 further comprises an optical band-pass filter 11 centered on
the wavelength of 370 nm; this optical band-pass filter 11 being arranged in front of
the diode 10.
20 For example, the optical band-pass filter 11 may be constituted of an optical
band-pass filter of reference "Biom-0007 Rev A - 370110" commercialized by the
Semrock company, having the following features:
- central wave length (CWL): 370 nm;
- full width at half the transmission peak (FWHM for Full Width at Half
25 Maximum): 11.5 nm + 1 nm;
- peak transmission percentage: 90% ;
- optical density (OD) > 5 for the 300-355 nm range ;
and
- optical density (OD) > 4 for the 385-1 000 nm range.
The curve C1 of fig.4 illustrates the transmission spectrum of such an optical
band-pass filter 11 according to the wavelength (in nm).
Of course, other pairs of diodes 10 and optical band-pass filters 11 may be
considered. In all cases, it is essential to select a diode by its spectral emission band
according to the spectral transmission band of the optical band-pass filter or
35 conversely, given that these two relatively narrow spectral bands must coincide.
In order to channel the radiation emitted by the diode 10 and filtered by the
optical band-pass filter 11, the system 1 comprises, behind the optical band-pass
filter 11, an objective lens 12, such as a convex plane lens particularly suitable for a
UV diode.
The system 1 further comprises, behind the objective lens 12, an optical
splitter 13 for splitting the main beam FP into a first sample-energizing beam FE and
5 a second reference beam FR. This optical splitter 13 is for example achieved in the
form of a semi-reflecting mirror or semi-reflecting prism or beam splitter.
For example, the optical splitter 13 may be constituted of a semi-reflecting
glass prism of BK7 type having a refractive (or transmission) rate of around 95 %,
and a reflecting rate of around 5%; the first energizing beam FE corresponding to the
10 refracted (or transmitted) beam by the optical splitter 13 and the second reference
beam FR corresponding to the beam reflected by the optical splitter 13. Fig. 5
illustrates the transmission spectrum of such a semi-reflecting glass prism of BK7
type-
As described previously, it is essential to recover a portion of the main beam
15 FP in the form of a second reference beam FR, to achieve the detection of the
intensity of the diode 10 and control its stability. Furthermore, the refraction (or
transmission) rate will be later taken into account in the determination of the energy
transmitted to the sample E.
The system 1 also comprises a support S for the sample E, this support S
20 being positioned so that the first energizing beam FE irradiates the sample E and
excites the 4-MU which will then emit a fluorescence ray RF as a result of the
excitation induced by this first energizing beam FE.
The fluorescence ray RF of the 4-MU (or fluorescence signal of the 4-MU) has
an emission peak at around 450 nm. The curve C11 of fig.7 illustrates the
25 fluorescence spectrum of the 4-MU which exhibits an emission peak at around 450
nm.
The system 1 comprises a first photodetector means 14 designed for detecting
this fluorescence ray FR, and outputting a first analog detection signal SAD1 in
response to a detection of this fluorescence ray RF. The first photodetector means 14
30 is for example of photodiode type, and the first analog detection signal SAD1
corresponds to a diode current (intensity in amperes).
For example, the first photodetector means 14 is constituted of a silicon
photodiode of reference "S1227 BR", commercialized by the Hamamatsu company,
and whereof the fig.6 illustrates the variation of the photosensitivity according to the
35 wavelength of the detected radiation, at a temperature of 25°C. Thus, it is to be
noted that such a silicon photodiode exhibits a photosensitivity in the magnitude of
0.25 AM/ for a wavelength in the magnitude of 450 nm, corresponding as a reminder
to the wavelength of the emission peak of the fluorescence ray RF of the 4-MU.
So that the first photodetector means 14 be sensitive to the fluorescence ray
RF of the 4-MU, the system 1 comprises, in a channel for guiding the fluorescence
ray RF placed between the sample E and the first photodetector means 14, an optical
band-pass filter 141 and a waveguide 142.
The optical band-pass filter 141 is substantially centered on the wavelength of
the fluorescence ray RF, and hence on the wavelength of 450 nm for the detection of
the fluorescence of the 4-MU.
For example, the optical band-pass filter 141 exhibits the following optical
features:
- central wavelength: 450 nm + 5 nm;
- width of the bandwidth: 40 nm + 4 nm;
- percentage of transmission at the central wavelength > 45% in the
magnitude of 80 to 90%.
The curve C13 of fig.7 illustrates the transmission spectrum of such an optical
15 band-pass filter 141, which exhibits a transmission peak coinciding substantially with
the fluorescence spectrum peak of the 4-MU illustrated on the curve CIA.
The waveguide 142 is preferably achieved in the form of a guiding cone or
optical cone, for example of polymethyl methacrylate (PMMA), in order to channel the
fluorescence ray RF filtered by the optical band-pass filter 141 in the direction of the
20 first photodetector means 14.
For example, the guiding cone 142 may exhibit a feature of the high-pass filter,
with a low-frequency cutoff lower than the wavelength of the fluorescence ray RF, in
this instance lower than 450 nm. The curve C12 of fig.7 illustrates the transmission
spectrum of such a guiding cone 142 which exhibits a low-wavelength cutoff in the
25 magnitude of 375-385 nm.
Fig.7 illustrates the fact that the guiding cone 142 does not affect the
fluorescence ray RF in spectrum terms, and that the selected optical band-pass filter
141 is suitable for detection by the first photodetector means 14. In fact, the optical
features of the optical band-pass filter 141 substantially coincide with the emission
30 spectrum of the fluorescence ray RF and highly reduce the interfering energizing
signals capable of being emitted by the irradiated sample E.
The system also comprises an objective lens 143 arranged in input of the
channel for guiding the fluorescence ray RF, before the optical band-pass filter 141 to
make the fluorescence ray RF converge towards the guiding cone 142; this objective
3s lens 143 may be of biconvex lens type.
Thus, the first energizing beam FE interacts with the medium contained in the
sample El thus causing the emission of a fluorescence ray RF which is collected by
the lens 143, before passing through the optical band-pass filter 141 and the guiding
cone 142 to the first photodetector means 14.
The system 1 also comprises a second photodetector means 15 designed to
detect the second reference beam FR, and output a second analog detection signal
5 SAD2 in response to a detection of this second reference beam FR. The second
photodetector means 15 is for example of the photodiode type, and the second
analog detection signal SAD2 corresponds to a diode current (intensity in amperes).
For example, the second photodetector means 15 is of the same type as the
first photodetector means 14 and can be constituted of a silicon photodiode of
10 reference 31227 BR" commercialized by the Hamamatsu company, and whereof
fig.6 illustrates the variation of the photosensitivity according to the wavelength of the
detected ray, at a temperature of 25°C. Thus, it is worth noting that such a silicon
photodiode has a photosensitivity in the magnitude of 0.17 ANV for a wavelength in
the magnitude of 370 nm, corresponding as a reminder substantially to the length of
15 the main beam FP, and hence of the second reference beam FR, after passing
through the aforementioned optical band-pass filter 11.
For the second photodetector means 15 to be sensitive to the second
reference beam FR, the system 1 comprises, in a channel for guiding the second
reference beam FR placed between the optical splitter 13 and the second
20 photodetector means 15, a low-pass filter 151 and a waveguide 152.
The optical low-pass filter 151 has a low-wavelength cutoff substantially lower
than the wavelength of the second reference beam FR, which as a reminder, is in the
magnitude of 370 nm.
For example, the optical low-pass filter 151 has the following optical features:
2 5 - transmission percentage in the magnitude of 40% + 5% for the wavelength
range ranging between 360 and 380 nm;
- transmission percentage lower than 1% for the wavelength range ranging
between 405 and 790 nm; and
- transmission percentage in the magnitude of 0.2% for the wavelength
30 range ranging between 425 and 790 nm.
The waveguide 152 is preferably achieved in the form of a guiding cone or
optical cone, for example polymethyl methacrylate (PMMA), in order to channel the
second reference beam FR filtered by the optical low-pass filter 151 in the direction of
the second photodetector means 15.
In the embodiment of fig.3 reflecting mirrors 144 and 154 are arranged at 45"
respectively after the objective lens 143 on the channel for guiding the fluorescence
ray RF and after the optical splitter 13 on the channel for guiding the second
reference beam FR, in order to be able to arrange the two photodetector means 14,
15 underneath the diode 10.
In other non illustrated embodiments, other optical members may be
incorporated such as for example additional lenses.
The second part of the following description relates to the processing part of
the signal of the system 1 in accordance with the invention which allows to analyze
the signals from the two photodetector means 14, 15 and to control the diode 10 in
order to carry out a detection by fluorometry of the analyte in the sample El with the
advantage of obtaining a very satisfactory signal-to-noise ratio.
Fig.8 schematically illustrates a system 1 in accordance with the invention
fitted with its signal processing part, according to a first configuration.
This signal processing part comprises three main boards, namely:
- a preamplification board 2 connected to the two photodetector means 14,
- a digital signal process board 3 arranged at the output of the
preamplification board 2 to calculate the quantity of analytes in the sample; and
- a driver board 4 of the diode 10 or "LED Driver Board" arranged at the
output of the digital signal process board 3 and driving the diode 10 in current.
The preamplification board 2 comprises:
- a first analogldigital conversion means 21 connected to the first
photodetector means 14 for converting the first analog detection signal SAD1 into a
first digital so-called fluorescence signal SFN.
- a second analogldigital conversion means 22 connected to the second
photodetector means 15 for converting the second analog detection signal SAD2 into
25 a second digital so-called reference signal SRN.
Fig.12 pertaining to a second configuration of the system 1, illustrates the
preamplification board 2 in a more complete and detailed manner.
As is visible on this fig.12, the first analogldigital conversion means 21
comprises:
30 - a first currentlvoltage converter 23 for converting the first analog detection
signal SAD1 (intensity in amperes) into a first detection voltage VD1 (in volts), with a
conversion of type VD1 = SAD1 .RDI, where RDI corresponds to a first resistance in
Ohms used for the current SAD1 /voltage VD1 conversion; and
- a first analogldigital converter 24 at the output of the first currentlvoltage
35 converter 23 for converting the first detection voltage VD1 into the first digital socalled
fluorescence signal SFN.
Similarly, the second analogldigital conversion means 22 comprises:
- a second currentlvoltage converter 25 for converting the second analog
detection signal SAD2 (intensity in amperes) into a second detection voltage VD2 (in
volts), with a conversion of type VD2 = SAD2.RD2, where RD2 corresponds to a
second resistance in Ohms used for the current SAD2lvoltage VD2 conversion; and
- a second analogldigital converter 26 at the output of the second
currentlvoltage converter 25 for converting the second detection voltage VD2 into the
second digital so-called reference signal SRN.
In an improvement illustrated on fig. 12, the first analogldigital conversion
means 21 further comprises a first subtracter 27 which subtracts from the first
10 detection voltage VD1 an offset voltage VOFFl in order to output a voltage VS1 = VOFFVDI,
this first subtracter 27 being interposed between the first currentlvoltage
converter 23 and the first analogldigital converter 24.
Similarly, the second analogldigital conversion means 22 further comprises a
second subtracter 28 which subtracts from the second detection voltage VD2 the
15 same offset voltage Voff, in order to output a voltage VS2 = Vo~~-VDt2h,is second
subtracter 28 being interposed between the second currentlvoltage converter 25 and
the second analogldigital converter 26.
The introduction of an offset voltage VoFF allows to use all the dynamic for
inputting digitallanalog converters 24, 26 by injecting in the latter voltages suitable for
20 their performance.
As illustrated on fig.8, the digital signal process board comprises a module for
generating modulationldemodulation signals 30, which comprises:
- a generator 300 outputting a sinusoidal carrier signal SNM at a predefined
carrier frequency fo, which is fixed for the rest of the description at 2 kHz; and
- an analog output module 301 arranged at the output of the generator 30.
With reference to fig.9, the sinusoidal carrier signal SNM is transmitted in
digital data packets of 16 bits, and comes in the form of a set of several periodic
sinusoidal iterations.
In the example of figs.9 and 10 (fig.10 illustrating one of the five iterations of
30 the three curves of fig.9), the number of iterations is four and each iteration is formed
of thirty one sinusoidal oscillations. Each sinusoidal oscillation (otherwise a unique
period of the sinusoidal signal) is constructed with forty sampling points each spaced
apart by 12.5 ps (namely 0.0125 ms).
Thus, the period of the sinusoidal signal inside an iteration is of 500 ps
35 (namely 0.5 ms, corresponding to forty times 0.0125 ms). Consequently, each
sinusoidal iteration lasts 15.5 ms, equivalent to thirty one times the period of 0.5 ms,
due to the thirty one sinusoidal oscillations in an iteration.
The time difference between two consecutive iterations, in other words
between the end of an iteration and the beginning of an iteration, is of 80 ms, much
higher than the period of sinusoidal oscillations (0.5 ms) and the duration of an
iteration (15.5 ms). Thus, the periodicity of the iterations is of 95.5 ms and in total, a
set of four iterations generally lasts 302 ms between the first and last iterations.
The generator 300 also emits two digital demodulation signals at this same
carrier frequency fo, namely:
- a digital demodulation signal SlNE at the carrier frequency fO and in phase
with the sinusoidal carrier signal SNM, this digital demodulation signal SlNE being
identical to the sinusoidal carrier signal SNM; and
- a digital demodulation signal COSINE at the carrier frequency fo and in
phase quadrature with respect to the sinusoidal carrier signal SNM.
For this, the generator 300 integrates a phase shifier at 90" for generating the
digital demodulation signal COSINE.
The analog output module 301, at the output of the generator 300,
successively comprises:
- a digitallanalog conversion means 303 for converting the sinusoidal carrier
signal SNM, before filtering, into an initial analog modulation signal SAM0 at the
carrier frequency fo; and
- at the output of the digitallanalog conversion means 303, an analog
processing module 304 which analogically processes the initial analog modulation
signal SAM0 for outputting an analog modulation signal SAM "or LED signal" at the
carrier frequency fo, this analog modulation signal SAM forming an analog drive
signal of the diode 10.
As illustrated on fig.12, this analog processing module 304 successively
comprises:
- a low-pass filter 305 of RC filter type;
- at the output of the low-pass filter 305, a subtracter 306 which subtracts
from the output voltage of the filter 305 a so-called offset voltage Txoffset, which
allows to maintain the diode 10 switched off when the initial analog modulation signal
SAM0 is zero, with for example Txoffset = 0.012 V; and
- at the output of the subtracter 306, an amplifier 307 of predefined gain,
and which outputs the analog modulation signal SAM.
Before going on to describe the digital signal process board 3, it is worth
noting that the driver board 4 comprises an amplitude modulator connected to the
digitallanalog conversion means 303 and to the diode 10 for modulating in amplitude
the main beam FP at the carrier frequency fo, by applying the analog modulation
signal SAM on the diode 10.
The driver board 4 more particularly comprises a voltagelcurrent converter 40
for converting the analog modulation signal SAM into a drive current ID of the diode
10 via a resistance RD in Ohms. The real voltage VD measured at the terminals of
the resistance RD corresponds to a sinusoidal feedback signal of the diode 10 (LED
Feedback Signal) and satisfies the relationship VD = ID.RD.
Thus, the driver board 4 drives in voltage and hence in intensity the diode 10
such that it emits a main beam FP composed of several flashes regularly spaced
over time, each flash being modulated at the carrier frequency fo, in other words
exhibits a sinusoidal component at the carrier frequency fo.
In the example of fig.12, the digital signal process board 3 further comprises a
retroactive control module 308 that receives in input the measured voltage VD
(sinusoidal feedback signal) coming from the driver board 4 in order to determine the
error or the shift between this voltage VD measured at the terminals of the resistance
RD and the analog modulation signal SAM which drives the diode 10, for a servocontrol
by return loop of the drive current ID of the diode 10.
This retroactive control module 308 successively comprises:
- at the output of the driver board 4, a summer 309 which sums up the
measured voltage VD and a so-called offset voltage RXoffset, which allows to
compensate for the incoming negative voltage VD when the initial analog modulation
signal SAMo is zero, with for example Rxoffset = 0.038 V;
- at the output of the summer 309, an amplifier 313 of predefined gain; and
- at the output of the amplifier 313, an analogldigital converter 31 1 which
outputs a digital feedback signal SFD.
A first purpose of this retroactive control module 308 is to be able to adjust the
amplitude and the zero (typically called "offset") of the drive current ID (ID = VDIRD),
so that this drive current ID corresponds to the required sinusoidal signal.
A second purpose of this retroactive control module 308 is to be able to
control, with each signal iteration, the shift between the drive current ID and the
analog modulation signal SAM, so as to generate alerts if a shift or error threshold is
exceeded.
Due to the modulation of the main beam FP, the two analog detection signals
SADI, SAD2, measured at the terminals of the two photodetectors 14, 15, also each
come in the form of a set of several periodic sinusoidal iterations at the carrier
frequency fo.
Thus, and as illustrated on figs. 9 and 10, the first digital fluorescence signal
SFN (or "Fluo signal") and the second digital reference signal SRN (or "Ref Signal"),
at the output of the preamplification board 2, each come in the form of a set of
several periodic sinusoidal iterations, substantially having the same temporal and
frequential features as the sinusoidal carrier signal SNM.
The digital signal process board 3 comprises a main signal
acquisition/processing module 31 arranged at the output of the generator 300 and the
5 preamplification board 2. This digital signal process board 3 comprises a first module
for acquiring/processing 32 the digital fluorescence signal SFN (or "Fluo Signal") and
a second module for acquiring/processing 33 the second digital reference signal SRN
(or "Ref Signal").
From a communication aspect, the digital signal process board 3 uses for the
10 transmission of the sinusoidal carrier signal SNM and the reception/acquisition, of the
first digital fluorescence signal SFN and the second digital reference signal SRN, a
digital communication bus BCO such as for example of SPI type (Serial Peripheral
Interface Bus) with 32 bits, capable of receiving and emitting 32 bits at the same
time.
With reference to fig.1 I, the generator 300 of the digital signal process board 3
emits at a regular interval a sampling point PEM(i) (i being an integer) for the
sinusoidal carrier signal SNM within a data packet PDll of 16 bits of the frame of 32
bits, the other data packet PD12 of 16 bits being unused. As a reminder, the time
difference between two successive sampling points PEM(i-1) and PEM(i) is of 12.5
20 ps (namely 0.0125 ms), which corresponds to a time difference between two
successive frames of 12.5 ps.
With reference to fig.1 I , each time a sampling point PEM(i) is transmitted for
the sinusoidal carrier signal SNM, the digital signal process board 3, and more
particularly its main signal acquisition/processing module 31, receives a sampling
25 point PEM(i-1) for the first digital fluorescence signal SFN and a sampling point
PEM(i-1) for the second digital reference signal SRN; the sampling points PEM(i-1)
and PER (i-I) being the responses to the previously emitted sampling point PEM(i-1)
for the sinusoidal carrier signal SNM. The sampling point PEF(i-1) and PER(i-1) are
transmitted and received in two respective data packets PD21, PD22 of 16 bits of the
30 frame of 32 bits.
Thus, the time interval between two sampling points PEM(i) for the sinusoidal
carrier signal SNM (12.5 ps) corresponds to the same acquisition time interval
between the sampling points PEF(i) and PER(i).
Thanks to the phase and quadrature coherent demodulation process
35 implemented by the main signal acquisition/processing module 31 and described
later, all the phase shifts possibly introduced by the system not affecting the signal
analysis for extracting a quantification value of the analyte in the sample E; the origin
of such phase shifts may be in response times between the drive current ID and the
emission of the main beam FP, delays introduced along optical paths of the beams
FP, FE, FR, RF, response times between the detection of the fluorescence ray RF by
the first photodiode 14 and the generation of the first analog detection signal SADI,
response times between the detection of the second reference beam FR by the
5 second photodiode 15 and the generation of the second analog detection signal
SAD2, the phase inversion introduced by the first subtracter 27 of the first
analogldigital conversion means 21 and the phase inversion introduced by the
second subtracter 28 of the second analogldigital conversion means 22 (see fig.l2),
trans-impedance amplification delays introduced by parasitic capacitors.
It is particularly interesting to generate a sinusoidal carrier signal SNM in the
form of a set of several periodic sinusoidal iterations.
In fact, it is possible to use the first iteration only to check if the two analog
detection signals SAD1 , SAD2 are above predefined minimum threshold, for example
fixed at -7V and +8V. If the first detection signal SAD1 is below the corresponding
15 minimum threshold, a return loop is provided for reducing the analog modulation
signal SAM, particularly in the magnitude of three times with respect to the initial
signal.
Thus, a correction coefficient is applied upstream on the module of the first
detection signal SADI, for each iteration, in order to compensate for a decrease of
20 the first energizing beam FE. The purpose of such a return loop is to prevent an
incorrect reading due to electronic channel saturations, and also to extend the
fluorescent reading range, such that even high concentrations of 4-MU may be
detected.
The three other iterations are used to extract the values of the modules of the
25 first digital fluorescence signal SFN and the second digital reference signal SRN, for
each iteration.
Inside each of these three iterations, the last sinusoidal oscillation is used to
check the voltage VD, also called the sinusoidal feedback signal of the diode 10 (Led
Feedback Signal) which corresponds as a reminder to the real voltage measured at
30 the terminals of the diode 10.
For the sake of precision, and in accordance with the amplitude modulation
principle, the module of the first digital fluorescence signal SFN corresponds to the
envelop of the sinusoidal part (inside a sinusoidal iteration) of the first digital
fluorescence signal SFN, and the module of the second digital reference signal SRN
35 corresponds to the envelop of the sinusoidal part (inside a sinusoidal iteration) of the
second digital reference signal SRN.
In the example illustrated on figs. 9 and 10, the digital fluorescence SFN and
reference SRN signals are modulated signals and the respective modules of these
two signals substantially correspond to half the voltage amplitude Vpp of the
sinusoidal part of the corresponding signal.
In order to detect a possible defect or a possible failure in the diode 10 or
optical devices 11, 12, 13 illustrated on figs. 2 and 3, the digital signal process board
5 3 automatically checks, at the end of each iteration, that the module of the second
digital reference signal SRN ranges within a predefined operating range, in other
words between two predefined safety thresholds.
The rest of the description specifically relates to the main signal
acquisitionlprocessing module 31, which allows to implement a phase and
10 quadrature coherent demodulation process.
The main signal acquisitionlprocessing module 31 includes, as a reminder:
- a first acquisitionlprocessing module 32 connected to the generator 300
and to the first analogldigital conversion means 21 of the preamplification board 2, for
acquiring and processing the first digital fluorescence signal SFN; and
- a second acquisitionlprocessing module 33 connected to the generator
300 and to the second analogldigital conversion means 22 of the preamplification
board 2, for acquiring and processing the second digital reference signal SRN.
The first acquisitionlprocessing module 32 comprises a first demodulation
means 34 designed to demodulate the first digital fluorescence signal SFN, and
20 comprising:
- a first multiplier 341 by the digital demodulation signal SINE in phase, this
first multiplier 341 being followed by a first low-pass filter 351 at a cutoff frequency fc
lower than the carrier frequency fo (for example fc = 110 Hz), in order to output a first
so-called in phase SFslN demodulated fluorescence signal; and
2 5 - a second multiplier 342 by the digital demodulation signal COSINE in
phase quadrature, this second multiplier 342 being followed by a second low-pass
filter 352 at the same cut-off frequency fc, in order to output a first so-called phase
quadrature SFcos demodulated fluorescence signal.
The first acquisitionlprocessing module 32 further comprises, at the output of
30 the first and second low-pass filters 351, 352, a means 353 for calculating a first socalled
fluorescence VALF value characteristic of the amplitude of the fluorescence
ray RF, corresponding to the module of the first digital fluorescence signal SFN and
thus to the module of the sum of the first in phase demodulated fluorescence s~gnal
SFSINa nd the first phase quadrature demodulated fluorescence signal SFcos.
More particularly, the first fluorescence value VALF is calculated according to
the following equation: VALF = (SFSIN2 + SFCO2S ) 1 12 .
The first acquisition/processing module 32 also comprises, at the input of the
first and second multipliers 341, 342:
- a notch filter 361 centered on the carrier frequency fo in order to filter the
first digital fluorescence signal SFN and output a first filtered intermediate signal
5 SFN'; and
- a subtracter 362 performing the subtraction of the first digital fluorescence
signal SFN and the first filtered intermediate signal SFN', in order to generate a first
filtered fluorescence signal SFNO = SFN - SFN', this first filtered fluorescence signal
SFNO being injected in the first and second multipliers 341, 342.
The second acquisition/processing module 33 comprises a second
demodulation means 37 designed for demodulating the second digital reference
signal SRN, and comprising:
- a first multiplier 371 by the digital demodulation signal SINE in phase, this
first multiplier 371 being followed by a first low-pass filter 381 at the cutoff frequency
15 f, lower than the carrier frequency fo in order to output a second so-called in phase
demodulated reference signal SRSIN; and
- a second multiplier 372 by the digital demodulation signal COSINE in
phase quadrature, this second multiplier 372 being followed by a second low-pass
filter 372 at the same cut-off frequency fcl in order to output a second so-called phase
20 quadrature demodulated fluorescence signal SFcos.
The second acquisition/processing module 33 further comprises at the output
of the first and second low-pass filters 381, 382, a means 383 for calculating a
second so-called reference value VALR characteristic of the amplitude of the
reference beam FR, corresponding to the module of the second digital reference
25 signal SRN and thus to the module of the sum of the second in phase demodulated
reference signal SRslN and the second phase quadrature demodulated reference
signal SRcos.
More particularly, the second reference value VALR is calculated according to
the following equation: VALR = (SRS~NS~RC' OS~)%.
30 The second acquisition/processing module 33 also comprises, at the input of
the first and second multipliers 371, 372:
- a notch filter 391 centered on the carrier frequency fo in order to filter the
second digital reference signal SRN and output a second filtered intermediate signal
SRN'; and
- a subtracter 392 performing the subtraction of the second digital reference
signal SRN and the second filtered intermediate signal SRN', in order to generate a
second filtered reference signal SRNo = SRN - SRN', this second filtered reference
signal SRNo being injected in the first and second multipliers 371, 372.
The rest of the description relates to the calculations implemented for
determining the fluorescence of the sample E by expressing it in the form of a relative
fluorescence unit (RFU), the relative fluorescence unit being defined as the ratio
between the intensity of fluorescence and the intensity of excitation.
The RFU values are usually adjusted for a required scale of values,
determined by a prior calibration process.
In the case of the present system 1, for a given concentration "xu of the 4-MU
contained in the sample E during measurement by fluorometry, the calibrated RFU
value is calculated as follows:
- Fx is a raw value in mV of the fluorescence signal detected in the first
photodetector, which corresponds within the scope of the invention, to the first
fluorescence value VALF (digital data) from the demodulation process implemented
in the digital signal process board 3;
- Rx is a raw value in mV of the reference signal detected in the second
photodetector, which corresponds within the scope of the invention, to the second
reference value VALR (digital data) from the demodulation process implemented in
the digital signal process board 3; and
- ~ F L U Oa nd REF are gain parameters adjusted during the optical calibration
of the system 1.
The calibration process is carried out by a prior analysis by fluorometry, of a
liquid solution of reference having a known concentration CREF of the 4-MU and
providing a given value of RFUREFlw ith for example CREF= 6410 nM and RFUREF=
3144 with a certain precision interval. The gain parameters ~FLUO and REF are
established during the analysis in order to obtain at the output a RFU in the
magnitude of RFUREF.
Within the scope of an immunological test, the concentration of the 4-MU is
situated in the 40 nM - 40000 nM range. Fig.13 illustrates the variation of the RFU
according to the concentration in 4-MU, with RFU non linear function of the
concentration in 4-MU. In order to obtain such a curve with a system 1 in accordance
with the invention, it is provided to apply at the output of the digital signal process
board 3 a conversion factor Fco~v, in this instance, a polynomial function of the
35 concentration in 4-MU, so that the RFU measured by the system 1 coincides with the
curve illustrated in fig.13 and thus obtain the required curve shape. This conversion
factor FcoNv is established by following an analysis protocol by fluorometry including
several variation factors (different solutions of 4-MU and different optical instruments)
in order to ensure the reliability and repeatability of the measurement.
The detail of the calculations is described hereinafter.
First, the drive current ID(t) of the diode 10 is established, from the conversion
5 of the analog modulation signal SAM. For a sinusoidal pulse, the drive current ID(t)
(or diode current) satisfies the following equation (El):
ID(t) = [A+B.sin (2Trfot)].rect~(t)n, amely
ID (t) = [A+B.cos (2Trfot- ~r12)rl.e ct~(t)
Where
- fo = 2 KHz (carrier frequency),
- T = 15.5 ms (duration of a sinusoidal pulse),
- A corresponds to the offset intensity (shift with respect to zero), which may
15 take the value of 15 mA, and
- B corresponds to the half-amplitude of the drive current, which may take the
value of 10 mA.
Fig.14 illustrates this drive current ID(t) according to time t for a pulse.
In the frequential field, this equation (El) is translated by the following
20 equation (E2):
Fig.15 illustrates ID(f) according to the frequency f.
25 Second, the drive current ID(t) (carrier signal from the modulation process at
the carrier frequency fo) is converted into the main beam FP at a given wavelength, in
this instance around 370 nm, before being split into a first energizing beam FE and a
second reference beam FR. The first energizing beam FE excites the sample and the
molecules of 4-MU, which in return emit a fluorescence ray RF detected by the first
30 photodetector 14; this photodetection being translated by a first analog detection
signal SAD1 at the terminals of the first photodetector 14.
It is established the following equation (E3) which translates the passage of
the drive current ID(t) to the first analog detection signal SADl(t), signal at the output
of the first photodetector 14 and at the input of the first analogldigital conversion
35 means 21 or the first analogldigital converter 24:
SAD1 (t) = z(t).lD(t), namely
SAD1 (t) = K.[A+B.cos(2Trfot - nl2 + A)].rectT(t), namely
SAD I (t) = K.A.rect~(t)+[KB. .c0s(2d0t+d)].rect~(t)
where
- z(t) is the amplification signal translating the emission of the fluorescence ray
5 RF, and which is considered as constant and equal to the constant K;
- a = A--rrl2, with A corresponding to a phase shift introduced by the system
between the diode 10 and the first photodetector 14.
Fig.16 illustrates this first analog detection signal SADl(t) according to time t
for a pulse.
Third, the analog detection signal SADl(t) is converted into a digital signal to
give the first digital fluorescence signal SFN, after passage in the first analogldigital
conversion means 21. In the frequential field, the following equation (E4) is thus
obtained:
SFN(f) = K.A.T.sinc(ff) + ~ . ~ . ~ 1 2 . ( d ~ . s i n c [ ( f+- efY~.)s~in)c][ (f+f0)T)]), namely
SFN(f) =Y(f) = K.A.T.sinc(ff) + ~.~.~12.(d~.sinc[(f-f~)~)]+e~~~.s(iEn4c)[ (f+f~)~)])
Fig. I 7 illustrates SFN(t) according to frequency f.
The first digital fluorescence signal SFN is then acquired and processed by the
digital signal process board 3, and more particularly by the main signal
acquisitionlprocessing module 31, according to the following steps of the
demodulation process.
In a first step of the demodulation process, the first digital fluorescence signal
SFN(f) passes through a band-pass filter composed of the notch filter 361 associated
with the subtracter 362.
The notch filter 361 translates into a function HnOtch(cf)e ntered on the carrier
frequency fo with a low-frequency cutoff fl = Ifol - fdec and a high-frequency cutoff f2 =
Ifol + fdec, where fdec can be set at 100 Hz, such that the frequential width Af notch of
the notch is Af notch = 2.fdec = 200 Hz. Fig.18 illustrates the function HNotch(af)c cording
to the frequency f.
The first filtered intermediate signal SFN1(f) hence satisfies the equation
SFN'(f) = HNotch(f).SFN(f).
Thus, the band-pass filter composed of the notch filter 361 associated with the
subtracter 362, translates by a function HB~(=~ 1) - HNotch(fw), hich is applied to the
first digital fluorescence signal SFN(f).
The first filtered fluorescence signal SFNo(f) = SFN(f) - SFN1(f), thus satisfies
the following equation (E5):
SFNo(f) = Hsp(f).SFN(f) = ( I - H~otch(f)).SFN(f), namely
SFNo(f) = ~.~.~/2.(d~.sinc[(f-+f, )e~-J)a] .sinc[(f+fo)~)])
Fig.19 illustrates SFNo(f) according to the frequency f.
In the temporal field, this equation (E5) translates by the following equation
SFNo(t) = [K.B.cos(2.nfot + B)].rectT(t), thus
SFN~(=~ )K .B.~~c~~(~).cos(~).co-s (K~.BT.&r~e)ct ~(t).sin(d).sin(2.nf~tn)a, mely
SFNo(t) = Kc(t).cos(2.nfot) - Ks(t).sin(2.nfot) (E6)
avec Kc(t) = K.B.rect~(t)c os(B) et Ks(t) = K.B.rect ~(t).sin(B)
In a second step of the demodulation process, the first filtered fluorescence
signal SFNO passes by the two multipliers 341, 342 so as to be multiplied by the
15 digital demodulation signals SlNE and COSINE. For the description of this step,
reference will be usefully made to fig.20 which specifically illustrates the two
multipliers 341, 342, as well as the low-pass filters 351, 352 and the calculation
means 353 that follows.
The SFNo(t) signal is multiplied, on the one hand, in the first multiplier 341 by
20 the digital demodulation signal SlNE so as to generate at the output a first in phase
intermediate demodulated signal SFSINl(t) and on the other hand, in the second
multiplier 342 by the digital demodulation signal COSINE so as to generate at the
output a first phase quadrature intermediate demodulated signal SFcosl(t).
The digital demodulation signals SlNE and COSINE come in the following
25 temporal forms:
SINE(t) = Ksin.~in(2.nfo=t) 2.sin(2.nf0t), and
COSINE(t) = K,os.cos(2~ot)= 2.cos(2rrf0t)
3 0 Where Ksin and Kc, correspond to the amplitudes of the signals and are set at
the value of 2.
The in-phase intermediate demodulated signal SFSIN~(ta) nd the phase
quadrature intermediate demodulated signal SFcosl(t) satisfy the following equations
(E7) and (E8), in the temporal field:
SFSINl(t)= [Kc(t).cos(2~rf~- tK) s(t).sin(2.nf0t) 1.2 sin(2rrf0t), namely
SFSIN1(t=) 2.[1/2.Ks(t) - 1/2.Kc(t).cos(2~~2f,t)1- /2.Ks(t).sin(2~~2f,t)],n amely
SFSIN8(=t) Ks(t) - Kc(t).cos(2~r2f~-t )K s(t).sin(2n2fot) (E7)
SFcosl(t) = [Kc(t).cos(21~f,t) - Ks(t).sin(2TTfot)1 .2 c 0 s ( 2 ~ ~ tn)a,m ely
SFcosl(t) = 2.[112.Kc(t) + 112. Kc(t).cos(2n2fot) - 112. Ks(t).sin(2n2fot)], namely
SFcosl(t) = Kc(t) + Kc(t).cos(2n2fot) - Ks(t).sin(2n2fot) (E8)
In the frequential field, these equations (E7) and (E8) translate into the
following equations (E9) and (E10):
SFSIN1(=f) [K.B.T.sin(d)].sinc(ff) -
[(K. B.T.cos(d))l2]. (sinc[(f-2fo)T)] + sinc[(f+2fo)T)]) -
[(K.~ .~.sin(d))l(2e]-.J TI'2.~inc[(f-2fo)d~")2]+.s inc[(f+2fo)~)]) (E9)
SFcosl(f) = [K.B.T.cos(d)].sinc(ff) +
[(K. B.T.cos(d))l2].( sinc[(f-2fo)T)]+sinc[(f+2fo)T)])-
[(K.B.T.sin(d))R] (e-J"12.sinc[(f-2fo)~+)] d TII2.sinc[(f+2fO)~)])
15 Fig.21 illustrates SFSIN1(f)a ccording to the frequency f and fig.22 illustrates
SFcosl(f) according to the frequency f.
In a third step of the demodulation process, the intermediate in phase
demodulated signal SFSIN1a nd the intermediate in phase quadrature demodulated
signal SFcos' are filtered by the respective low-pass filters 351, 352, which deliver at
20 the output respectively a first in phase demodulated signal SFSIN and a first in phase
quadrature demodulated signal SFcos.
Each low-pass filter 351, 352 is translated by a function HLp(f) with a cut-off
frequency fc set here at 11 0 Hz.
In a non illustrated improvement, each low-pass filter 351, 352 is followed by a
25 smooth low-pass filter which is translated by a function HLPsmoothil(lfu)s trated in fig.23,
with cut-off frequencies at fcl = 2fo, fc2 = 4fo, etc. Each smooth low-pass filter allows
to carry out an average out of the last twenty sampling points from the previous lowpass
filter HLp(f) to increase the rejection of the signals at the cut-off frequencies
multiple of 2fo.
3 o The first in phase demodulated signal SFslN and the first in phase quadrature
demodulated signal SFcos satisfy the following equation ( E l l ) and (E12), in the
frequential field:
SFSIN(f)= SFSIN'(f).HLP(f).H~Psmo=ot [hK(f.)B .T.sin(d)].sinc(fl) (El I )
SFcos(f) = SFcos1(f).H~p(f).H~prnth=(f )[ K.B.T.COS(~s)in] c(fT) (El21
Figs. 24 and 25 respectively illustrate signals SFsl~(f)andS Fcos(f) according to
the frequency f.
In the temporal field, these equations (El I ) and (E12) are translated by the
following equations (E13) and (E14):
In a fourth step of the demodulation process, the first in phase demodulated
signal SFslN and the first in phase quadrature demodulated signal SFcos are injected
in the calculation means 353, in order to calculate the module of the first analog
10 detection signal SADl(t), which corresponds to the module of signal [(SFSIN(t)+
SF^^^(^))^]'"; these steps of the demodulation process allowing to extract the
amplification signal z(t) without being affected by the unknown phase shift introduced
by the system.
The calculated module, corresponding to the first fluorescence value VALF
15 generated at the output of the main signal acquisition/processing module 31, satisfies
the following equation (El 5):
VALF = Module [SADl(t)], namely
VALF = [ ~ c ( t+) ~Ks (t)2 ] 112
VALF = [(K.~ .rect~(t).cos(a+) )[~K .B .rect~(t).sin(a2) 1]1 12
VALF = [(~.~.re~tT(t))~.(cos~in((d8))2 ~) ]+1 12
VALF = K. B. rectT(t)
And hence, considering that rectT(t) = 1 fort < T
VALF = K.B (El51
The same equations are repeated for the reference beam FR, and thus for the
calculation of the second analog detection signal SAD2, of the second digital
reference signal SRN and at last the second so-called reference value VALR.
Fourth, a comparison means (not illustrated) arranged at the output of the first
30 acquisition/processing module 32 and the second acquisition/processing module 33
performs the calculation of the value of the RFU based on the first fluorescence
values VALFjSi and the second reference values VALR j,i delivered at the output of
these modules 32, 33, for each sinusoidal iteration (j corresponding to the number of
iterations and taking values 0, 1, 2, as, for a reminder, three iterations are used for
35 measuring fluorescence) and for a concentration in 4-MU numbered i.
For an iteration j, with a concentration i, the prior value of RFU, named rf~,,~,
satisfies the following equation (E16):
Where Xj,i satisfies the following equations:
- in normal coordinates:
((VALF,, - offset,,) K:?))
x.. =
J .I (VALR,, . A:;:) KSTRI.P K STDI
- in extended coordinates:
-
with
- i = number of the concentration in 4-MU
- j = number of the iteration (0, 1, 2)
Kz= a djustment coefficient for the fluorescence ray calculated during
the prior calibration process
= adjustment coefficient in extended coordinates for the
1s fluorescence ray calculated during the prior calibration process
MI:
- Am = adjustment coefficient for the reference beam calculated during the
prior calibration process
ICF
- Cupt = adjustment coefficient in extended coordinates for the reference
beam calculated during the prior calibration process
2 0 - K,sr~ = adjustment coefficient of internal reading calculated during the
auto-calibration
- K , , , = adjustment coefficient of reference optical position calculated
during the optical calibration process
- VALF,,, = first fluorescence value for the iteration j, obtained for the
2s concentration i
- VALR ,,, = second reference value for the iteration j, obtained for the
concentration i
- *fier=~firs,t mean fluorescence value obtained for a reading in the air
- = first mean fluorescence value in extended coordinates obtained
for a reading in the air
- rfu,,, = value of offset RFU
Then, after having calculated the j values of rfuj,i, a polynomial conversion is
performed as explained above, with reference to fig.13, relating to the conversion
factor Fco,qv.
For an iteration j, with a concentration i, the value of RFU after applying the
conversion factor Fco,qv,n amed rfu,.,C ONY , satisfies the following equation (E17):
CONV - flu -(fluJ,i)4'a+(,f,j,,)3'b+(rfu,,i)2'c+(rf~,,,)'d+(rfu,,,()E21'e7 )
Where a, b, c, d and e are constants, with for example:
- a = 1,3240745951 716500 E-13 ;
- b = -3,768670701 8928200 E-09 ;
- c = 7,3337036404781 100 E-07 ;
- d = 1,031 1832028790600 E+OO ;
- e = 9,4239190294182200 E-01 ;
and where:
- for a value of rfu~calculatedw ith (E17) and higher than a threshold
Rmax, here equal to the value of 23433, r f u y v is set at the threshold value Rmax;
and
- for a value of rfu~calculatedw ith (E17) and lower than the threshold
('ONV 25 Rmax, rfic,,, remains equal to this calculated value.
Finally, the value of the required final RFU R F U ~ ~fo'r~ a, concentration i, is
calculated by performing the average of rjkYv with j taking the values 0, 1 and 2
without possibly taking into consideration a value of rfu(,'O,N,G ' too far from the average.
At the end, the final equation (E18) is obtained:
Fifth, this value of the final RFU R F U ~is~ tr'a~ns mitted to an operating system,
such as an outer computer terminal or a computer system integrated within an
automated in vitro diagnosis instrument, for evaluating, with additional calculations,
the concentration in analytes.

We claim:
1. A system (1) for the in vitro detection andlor quantification by
fluorometry of at least one analyte in a sample (E) of fluid, comprising:
- a radiation source (10) emitting a main beam (FP) in a given wavelength
called emission wavelength;
- an optical splitter (13) arranged at the output of the radiation source (10)
for splitting the main beam (FP) into a first sample (E)-energizing beam (FE) and a
second reference beam (FR);
- a first photodetector (14) means designed for providing a first analog
detection signal (SADI) in response to detecting a fluorescence ray (RF) emitted by
the sample (E), in a so-called fluorescence wavelength as a result of the excitation
induced by the first energizing beam (FE);
- a second photodetector (15) means designed for providing at the output a
15 second analog detection signal (SAD2) in response to a detection of the second
reference beam (FR);
said system (1) being characterized in that it comprises:
- a generator (300) outputting a sinusoidal carrier signal (SNM) at a
predefined frequency called carrier frequency (f,), and at least one digital
20 demodulation signal (SINE, COSINE) at this same carrier frequency (f,);
- a digitallanalog conversion means (303) connected to the generator (300)
for converting the sinusoidal carrier signal (SNM) into an analog modulation signal
(SAM) at the carrier frequency (fo);
- an amplitude modulator (4) connected to the digitallanalog conversion
25 means (303) and to the radiation source (10) to modulate in amplitude the main
beam (FP) at the carrier frequency (fo) by applying the analog modulation signal
(SAM) on said radiation source (10);
- analogldigital conversion means (21, 22) connected to the photodetector
means (14, 15) to convert the first analog detection signal (SADI) into a first digital
30 so-called fluorescence signal (SFN) and the second analog detection signal (SAD2)
into a second digital reference signal (SRN);
- digital processing means (32, 33) connected to the generator (300) and to
the analogldigital conversion means (21, 22), designed, on the one hand to process
the first digital fluorescence signal (SFN) by demodulation at the carrier frequency (fo)
3s in order to calculate a first so-called fluorescence value (VALF) characteristic of the
amplitude of the fluorescence ray (RF) and, on the other hand, process the second
digital reference signal (SRN) by demodulation at the carrier frequency (fo) in order to
calculate a second so-called reference value (VALR) characteristic of the amplitude
of the reference beam (FR);
- a means for comparing the first fluorescence value (VALF) and the second
reference value (VALR) to calculate a final result (RFU) for establishing the detection
5 andlor quantification of the analyte.
2. The system (1) according to claim 1, wherein the digital processing
means (32, 33) comprise:
- a first demodulation means (34) designed for demodulating the first digital
fluorescence signal (SFN) by multiplying it by at least one digital demodulation signal
10 (SINE, COSINE) at the carrier frequency (fo), in order to generate at least one first
demodulated fluorescence signal (SFSIN,S FCOS)a;n d
- a first calculation (353) means designed to calculate, based on at least
one first demodulated fluorescence signal (SFSIN, SFCOS), one first so-called
fluorescence value (VALF) characteristic of the amplitude of the fluorescence ray
15 (RF);
3. The system (1) according to claim 2, wherein the first demodulation
means (34) comprises:
- a first multiplier (341) by a digital demodulation signal (SINE) at the carrier
frequency (fo) and in phase with the sinusoidal carrier signal (SNM), said first
20 multiplier (341) being possibly followed by a first low-pass filter (351) at a cut-off
frequency (f,) lower than the carrier frequency (fo), in order to output a first so-called
in phase demodulated fluorescence signal (SFSIN); and
- a second multiplier (342) by a digital demodulation signal (COSINE) at the
carrier frequency (fo) and in phase quadrature with respect to the sinusoidal carrier
25 signal (SNM), said second multiplier (342) being possibly followed by a second lowpass
filter (352) at the same cut-off frequency (f,), in order to output a first so-called
phase quadrature demodulated fluorescence signal (SFcos);
and wherein the first calculation means (353), at the output of said first and
second multipliers (341, 342), calculates the first fluorescence value (VALF)
30 corresponding to the module of the sum of the first in phase demodulated
fluorescence signal (SFSIN) and the first phase quadrature demodulated fluorescence
signal (SFcos).
4. The system (1) according to claim 3, wherein the digital processing
means (32, 33) comprise, at the input of the first and second multipliers (341, 342):
- a notch filter (361) centered on the carrier frequency (fo) in order to filter
the first digital fluorescence signal (SFN) and output a first filtered intermediate signal
(SFN'); and
- a subtracter (362) performing the subtraction of the first digital
fluorescence signal (SFN) and the first filtered intermediate signal (SFN'), in order to
generate a first filtered fluorescence signal (SFNO), this first filtered fluorescence
signal (SFNO) being injected in said first and second multipliers (341, 342) of the first
demodulation means (34).
5. The system (1) according to any one of the preceding claims, wherein
the digital processing means (32, 33) comprise:
- a second demodulation means (37) designed to demodulate the second
digital reference signal (SRN), by multiplying it by at least one digital demodulation
signal (SINE, COSINE) at the carrier frequency (fo), in order to generate at least one
second demodulated reference signal (SRSIN,S RCOS)a; nd
- a second calculation means (383) designed to calculate, based on at least
one second demodulated reference signal (SRSINI SRCOS)t,h e second reference
value (VALR).
6. The system (1) according to claim 5 wherein the second demodulation
means (37) comprises:
- a first multiplier (371) by a digital demodulation signal (SINE) at the carrier
frequency (fo) and in phase with the sinusoidal carrier signal (SNM), said first
multiplier (371) being possibly followed by a first low-pass filter (381) at a cutoff
20 frequency (f,) lower than the carrier frequency (fo), so as to output a second so-called
in phase demodulated reference signal (SRSIN); and
- a second multiplier (372) by a digital demodulation signal (COSINE) at the
carrier frequency (fo) and in phase quadrature with respect to the sinusoidal carrier
signal (SNM), said second multiplier (372) being possibly followed by a second low-
25 pass filter (382) at the same cutoff frequency (f,) so as to output a second so-called
phase quadrature demodulated reference signal (SRcos);
and wherein the second calculation means (383), at the output of said first and
second multipliers (371, 372), calculates the second reference value (VALR)
corresponding to the module of the sum of the second in phase demodulated
30 reference signal (SRSIN) and the second in phase quadrature demodulated reference
signal (SRcos).
7. The system (1) according to claim 6 wherein the digital processing
means (32, 33) comprise, at the output of the second analogldigital conversion
means (22) and at the input of said first and second multipliers (371, 372):
3 5 - a notch filter (391) centered on the carrier frequency (fo) so as to filter the
second digital reference signal (SRN) and output a second filtered intermediate
signal (SRN'); and
- a subtracter (392) performing the subtraction of the second digital
reference signal (SRN) and the second filtered intermediate signal (SRN') so as to
generate a second filtered reference signal (SRNo), this second filtered reference
signal (SRNo) being injected in said first and second multipliers (371, 372) of the
second demodulation means (37).
8. The system (1) according to any one of the preceding claims, further
comprising, interposed between the radiation source (1 0) and the optical splitter (I3) ,
an optical band-pass filter (1 1) centered substantially on the emission wavelength.
9. The system (1) according to any one of the preceding claims, further
comprising, interposed between the sample (E) and the first photodetector means
(14), an optical band-pass filter (141) centered substantially on the fluorescence
wavelength.
10. The system (1) according to claim 9, further comprising, interposed
between the optical band-pass filter (141) and the first photodetector means (14), a
waveguide (142) achieved in the form of a guiding cone.
11. The system (1) according to any one of the preceding claims, further
comprising, interposed between the optical splitter (1 3) and the second photodetector
means (15), an optical low-pass filter (151) which exhibits a low-wavelength cutoff
substantially lower than the emission wavelength.
12. The system (1) according to claim 11, further comprising, interposed
between the optical low-pass filter (151) and the second photodetector means (15), a
waveguide (1 52) achieved in the form of a guiding cone.
13. The system (1) according to any one of the preceding claims, wherein
the sinusoidal carrier signal (SNM) is in the form of a set of several periodic
sinusoidal iterations at the carrier frequency (fo), the time difference between two
consecutive iterations being higher than the period of sinusoidal iterations.