The invention relates to methods for measuring the concentrations of particles suspended in a liquid sample, namely, to coherent fluctuation nephelometers (hereinafter - CFN) and CFN based microbiological analyzers used for registration of microbiological flora growth in biological samples in a wide range of turbidity values and for diagnostics and morphological studies of suspended microbial particles, in particular.

The invention also relates to immunochemical and hematological analyzers applied for obtaining information not only on absolute values of concentration in a wide range of values, but also, on morphological parameters of light scattering particles and transformation of such parameters both at low and high concentrations of particles. Registration of immunological reactions associated with aggregation and disaggregation of particles in the processes of latex agglutination, transformation or cell lysis, could serve an example of such biomedical applications

There are three main ways to determine the concentration of particles suspended in a liquid sample by recording the turbidity of suspended matter: turbidimetry (hereinafter - T), nephelometry (hereinafter - N) and CFN. All these methods are based on a general principle, namely, on recording the intensity and / or fluctuations of the intensity of transmitted or scattered light detected by one or more detectors upon the elastic scattering of light by particles in the liquid sample.

T is used for determining the intensity of light transmitted through the studied sample, however, this measurement is characterized by low sensitivity. The main disadvantage of N is that for achieving high sensitivity there is a need exclude any parasitic light from optical elements and transmitted beam on the detector. That is why in multi-angle N apparatus, which are used for measuring distribution of intensity of scattered light by the angles, transition to small angles is coupled with the need of solving the problem of parasitic light exclusion. It becomes possible only under conditions of large distance between cuvette and detector, high quality of cuvettes and other elements of the optical path. Therefore, multi-angle N cannot be used to measure concentrations of particles in a wide range of values; they are constructively complicated, and cannot be used as a basis for multichannel systems . [Clin Chem Lab Med. 2012 Jul;50(7): 1253-62. Low angle light scattering analysis: a novel quantitative method for functional characterization of human and murine platelet receptors. Mindukshev I etal.]

Dynamic ranges of turbidity measurements for each of these methods are different. Particularly, to measure bacterial concentrations in the colony-forming units (hereinafter - CFU), these methods have corresponding dynamic ranges: for T it is 10 ⁷ -10 ¹⁰ CFU / ml, for N it is 10 ⁵ -10 ⁸ CFU / ml, and for CFN it is 10 ³ -10 ⁸ CFU/ml. It is important that in many applications concentrations and morphology of particles suspended in liquids must be carried out in a wide range of turbidity, as well as, in a wide range of morphological parameters of the suspended particles.

Often the measurement process has to be interrupted due to the need of switching to an optical scheme with a reduced or enlarged length of the optical path of the reference light beam, which can be achieved by changing the geometric size of the optical cuvette thus obstructing analytical laboratory diagnostics processes.

Another widely applied analytical approach to determine the concentration and morphology of particles suspended in a liquid sample in a wide range of turbidity is dilution of the suspension in a given optical cuvette and in a given optical instrument. However, in a complex process the mentioned above procedures of dilution or changing size of the cuvettes will affect the course of the studied processes itself, whose efficiency (for example, the flow rate) depends on concentration. Besides, quite often researched system turns out to be unique and difficult to be reproduced, any interference in the observed process leads to its destruction. Moreover, any extra interference in the microbiological study can lead to contamination of samples by extraneous microorganisms and distortion of its results. Thus, the expansion of the dynamic range of measured concentrations under the given observation conditions for studied multi-parameter systems is definitely actual.

Measurement of the light scattered at small angles using the N and CFN methods is more sensitive. However, it has the disadvantage that makes it not applicable for measuring turbidity at high concentrations of scattering particles, as the intensity of the detected signal with increasing concentration at definite moment has ceased to depend on concentration of scattering particles due to the multiple scattering effect and with a further increase in concentration, the signal starts to decrease. Thus, the dependence of the detected signal for N and CFN is nonmonotonic in a wide range of angles. The T method, which is based on measuring the intensity of the non-scattered part of the reference beam of the light, allows to detect high turbidity in a monotonic regime; however, it is not sensitive to low concentrations of scattering particles, i.e. its signal practically does not change in this concentration range.

There are apparatus, which combine N and T measurements in one appliance to expand the dynamic range of turbidity measurements. For example, there is a specific proteins analyzer IMMAGE800 (Beckman Coulter), which uses two separate detectors to record the N signals with an angle of 90 degrees and T simultaneously. Such devices are convenient, for example, for registering concentration of protein and colloid particles of nanometer size, because, such particles with a size much smaller than the wavelength of visible light, scatter the light almost uniformly over all angles, while the nephelometric registration of the scattered light at a large angle can reliably separate the light scattered by these particles from stray light caused by the direct beam scattering mainly on the defects of optical cuvette.

At the same time, the turbidity detection by means of these devices in a wide dynamic range due to larger particles, for example, bacteria with a size comparable to visible light, becomes less effective, because the large particles scatter the light mainly forward, thus, N recording at large angles becomes inefficient. A combination of T and N at small angles is extremely difficult. However, the recording of turbidity for particles larger or comparable to the wavelength of scattered light is of great practical interest.

There is a microbiological analyzer HB&L (Alifax S.p.a.), which records the intensity of scattered light at several angles to increase the efficiency of N measurement of the turbidity caused by a bacterial culture, which makes it possible to record the growth of a bacterial culture more reliably. Meanwhile, the dynamic range of measured concentrations remains within the above-mentioned limits inherent in the N method. Eventually this leads, among other things, to the fact that dirt on the cuvette and its defects can differently affect both signals, which results in deterioration of the reproducibility of measurements.

Thus, the drawbacks of the reviewed methods are either insufficient sensitivity due to their sensitivity to parasitic stray light and / or limited dynamic range of different recording channels, which leads to the need to use a variety of devices and measurement conditions for determination of turbidity in a wide range of values.

However, there are tasks in which you need to monitor the processes with changing turbidity in a wide range of values, for example, while registering growth curves of microorganisms in a nutrient medium, including the ones with antibiotics, which are implemented in biotechnological and diagnostic systems related to the growth of microflora, as well as, in case of registration of latex aggregation or disaggregation when the size of the particles is comparable to or larger than the reference light.

None of the reviewed methods allows to combine the processes of detection of turbidity (i.e., intensity of scattered and / or transmitted light) and morphology of the particles during the aggregation, disaggregation, division, lysis, etc., which registration is also complicated by the changing polydispersity of the researched systems. The ability to control the turbidity and morphology of particles simultaneously in a wide range of turbidity and morphology creates the new possibilities for studying and identifying complex structures in liquids and specifically the ones of the living systems.

There is a known method of measuring concentration of particles in a wide range of turbidity [Scientific Journal: Apparatus and Experiment Technique. Scientific article on the theme: "Coherent Fluctuation Nephelometry: Highly Sensitive Method of Detecting Particles in Liquids". Author: Rastopov S. F.], where the reference beam of coherent radiation passes through a cuvette with the studied sample when the intensities of direct transmitted radiation and scattered radiation being recorded with the help of photodetector devices (PDDs) and the intensity of the direct radiation transmitted through the cuvette with the tested sample T (turbidimetry) and fluctuations in the intensity of scattered light intensity fluctuations (SLIF) are measured at two scattering angles. This method allows to achieve measurement sensitivities up to 10 ³ - 10 ⁴ particles/ml with a quite simple appararus design.

The disadvantage of the prototype is that this method of detecting micron and submicron particles in a liquid is based on measuring the fluctuations in the intensity of coherent light scattered by particles and it does not allow evaluating morphology of the particles. It is not possible to measure the frequency characteristics of SLIF with this method.

The closest to the claimed method is the analyzer [DOI: HTTP://D .DOI. _ORS/IO .I5 _769/222O -7619-2OI6-4-39S- ₃₉₆ ], which simultaneously uses one CFN registration channel and one T channel to expand the dynamic range, with the viewing angle chosen to be optimal for bacterial cells with the size of about 1 pm and selected concentration range (10 ⁴ -10 ⁸ CFU / ml for CFN ) of about 7 degrees.

The disadvantages of the prototype are as follows:

- the signal recording in the prototype is nonmonotonic in a wide dynamic range of measured turbidity and morphological parameters of particles, i.e. particle sizes that can be analyzed are limited;

- impossibility of detecting scattered light at different angles, which is necessary to optimize the detection of particles in a wide range of sizes; WO 2019/143271 _, PCT/RU2019/000170. . .

- prototype does not allow to determine the morphology ot particles, which is important, for example, for particle aggregation and changes in the structure of blood cells.

- it is also difficult to separate the variations in the signals of CFN and T, caused either by a change in the number of particles or by the change in their morphology.

The aim of the claimed invention is to expand the dynamic range of measured turbidity and evaluation of morphological composition of particles in the tested sample, which is achieved by simultaneous recording of T, SLIF at different scattering angles (at least two), calculating a linear combination of SLIF initial signals at different scattering angles and a signal of the T turbidimeter preferably within degree ranging from 2 to 4 with different weight ratio, which are determined by calibration for the suspension of analyzed particles of each type. By means of calculated linear combination of signals, one can estimate the concentration of particles, using correlation of SLIF signals at different angles it becomes possible to define the morphology of particles in the sample. The dependence of such resulting signal on the particle concentration is monotonic in a wide range of concentrations and particle morphologies. The registration of SLIF signals simultaneously at several angles allows detecting particles of different sizes with maximum sensitivity. The registration of the SLIF signal correlation allows evaluating particle morphology, for example, in particle aggregation processes.

Additionally, the frequency spectra of SLIF are measured in the frequency range of more than 1 Hz and T frequency spectra - preferably in the frequency range of less than 1 Hz in order to separate the changes in particle morphology from changes in their number, as well as, to detect the effect of large contaminants on the resulting signal. These frequency limits are optimized on the basis of experimental data.

The device, according to the claimed method (an embodiment example), comprises a source of coherent radiation, made in the form of a laser, cuvette with the tested sample, and direct, transmitted and scattered radiation intensity detector (registering device) made in the form of photodetector devices (PDDs), characterized by three or more PDDs located at different angles to the axis of the laser beam, while the first PDD is located on the axis of the radiation beam (zero angle), and its output is connected to the input of the intensity meter, while the outputs of remaining PDDs are connected to the inputs of the intensity fluctuation meters through the suppressor of the signal's constant component, where the input of the intensity meter and outputs of intensity fluctuation meters are connected to calculation device, which calculates the linear combination of SLIF signals and T signal preferably in the degree ranging from 2 to 4.

In order to separate the changes in the morphology of particles from changes in their number, as well as, to detect the effect of large contaminants on the resulting signal, the device also includes the T frequency spectrum meter, working preferably in a frequency range less than 1 Hz, and SLIF frequency spectrum meter working preferably in a frequency range of more than 1 Hz.

In order to increase the stability to fluctuations in the intensity of the radiation source, PDD recording SLIF are preferably made in the form of two photosensitive elements (PSEs) for each scattering angle positioned symmetrically from the axis of the laser beam, the outputs of these PSEs are connected to various inputs of the respective differential amplifiers.

To detect particles of different sizes in order to optimize detection angles, the cuvette and / or PDD are arranged so that they can be moved along the axis of the radiation beam.

The invention is explained by seven figures.

Picture 1 is a block diagram of an apparatus according to the claimed method, wherein 1 is a laser; 2- cuvette with the tested sample; 3 - axis of the laser beam, 4 - PDD of T signal; 5.6 - PDD of SLIF signals; 7 - calculation device (CD); 8 - T signal intensity meter; 9,10 - SLIF signal intensity meters; 11, 12 - suppressors of signal's constant component. Picture 2 is a block diagram of subtracting SLIF signals at a given scattering angle (device’s top view), where 5a and 5b are various photosensitive elements (PSEs) of PDD.

Picture 3 shows experimental data of recording the growth of E. coli bacteria, where 13 - T signal; 14 - SLIF signal in one scattering angle; 15 - resulting signal according to formula (1) below for this angle.

Picture 4 schematically shows the resulting signals (1) at different scattering angles without changing the particle size during growth (solid lines) and aggregation (dashed lines).

Picture 5 presents experimental data on the growth registration and

"clumping" of bacteria Staphylococcus aureus, 16 - T signal; 17 - SLIF signal at one scattering angle.

Picture 6 shows the experimental SLIF frequency spectra at one scattering angle for measuring the turbidity of a suspension of particles without contaminants (on the left) and during the pass of a large foreign particle through a radiation beam (on the right).

Picture 7 shows the experimental SLIF signals at four different scattering angles for the red blood cell suspension, which sediment and aggregate over time.

The essence of the claimed method for determining the concentration and morphology of suspended particles in a wide range of turbidity, is the passage of a probing beam of coherent radiation through a cuvette with the test sample, registration the intensities of the direct transmitted and scattered radiation by means of photodetector devices (PDDs) with simultaneous measurements of the T and SLIF intensity of the light transmitted through the cuvette with the test sample at several scattering angles . By the intensity ratio of SLIF at different angles, it is possible to assess the size of the analyzed particles by means of the scattering indicatrix. For example, as the particle size increases over time (agglutination), the scattered light is concentrated mainly at smaller angles, which is determined by the signals ratio at different angles. To provide the monotonicity of the output signal, it is necessary to calculate a linear combination of the SLIF initial signals at different angles and T signal of the turbidimeter, taken within the degree from 2 to 4, with different weight coefficients, which are determined by calibration for suspensions of the analyzed particles of each type.

The SLIF initial signals at different angles and T can be summed according to the formula:

wherein, If ^UF - rms deviation of SLIF signal at PDD with number i, - Tsignalintensity, ^ff i - weight coefficients, £-degreepower coefficient within the range from 2 to 4.

Thus, the generated combination of signals allows to expand the dynamic range of the measured concentrations, as the resulting signal monotonically depends on the particle concentration. This summing of SLIF signals also leads to an increase in the sensitivity of measurements at low concentrations. In addition, the resulting signal indicates on the concentration of scattering particles of this type, while the ratio of SLIF signals at different angles specifies the particle sizes.

For estimation of the absolute concentration of particles, it is necessary to carry out a preliminary calibration of the resulting signal on the concentration for specific sample particles of different morphology.

A real sample, in addition to the studied particles, may contain different contaminants of a relatively large size, as well as air bubbles that distort the T and SLIF signals. To reduce the effect of such extraneous particles on the measured concentration of the analyzed particles, the frequency spectra of the signals are also calculated. The registration of the SLIF signal spectra allows increasing the reliability and accuracy of the measurement. For example, when a beam intersects with large contaminants, or when air micro bubbles are formed on the walls of a cuvette, the standard shape of the Fourier spectrum of the SLIF signal is disrupted. It has been experimentally shown that such violations for T appear preferably at the frequencies below 1 Hz, while for SLIF - preferably at the frequencies above 1

Hz. And, as a rule, in these cases the signal is not violated from all PDDs located at different angles. In such measurements, signals with a disturbed Fourier spectrum can be excluded from the resulting signal, thereby reducing the influence of contaminants and bubbles. In addition, in some processes, for example, when registering the growth of Staphylococcus aureus bacteria, the total number of bacteria continuously increases, but the signals of SLIF and T begin to decrease.

This is being caused by massive aggregation of bacteria, known as "clumping”.

The increase in the low-frequency fluctuations of the T signal makes it possible to register this process and distinguish it from the death of the bacterial culture, which also leads to a fall in the SLIF and T signals, but without increasing the low- frequency fluctuations of the T signal. Thus, the proposed method and apparatus allow collecting multiparametric information on the processes occuring in the suspended particles. This information is related to the number of particles (as determined by the SLIF and T signals), with their morphology (as determined by the ratio of SLIF signals at different angles), and dynamics of particle motion in the cuvette (as determined by the Fourier spectral distributions of the SLIF and T signal). Such a multiparametricity makes it possible to control the processes occurring in the researched particle suspension in form of pattern recognition formed by the neural network and thus allowing to identify the type of particles participating in the processes, as well as, their number.

The apparatus implemented according to the method is as follows (an example of a particular preferred implementation, Picture 1). The device consists of a source of coherent radiation, made in the form of a laser, 1 , a cuvette with a analyzed sample 2, located on the axis of the laser beam 3, several photodetector devices (PDDs) for recording the T signal 4 and SLIF signals 5, 6 (picture shows the first two, while the total number (typically 8) depends on the required detailing of measurements at different angles) and a calculation device (CD) 7. In this case, the output of PDD 4 is connected to the input of the intensity meter 8 (for example, an operational amplifier (OA)), while the outputs of remaining PDDs (5 and 6) are connected to the inputs of the intensity fluctuation meters 9 and 10 (also OA) through the suppressors of signal's constant component 11 and 12 (in the simplest case, these are simply the separation capacitors). The outputs of all OAs are connected to the inputs of CD 7. A multichannel analog-to-digital converter (ADC) and computer or microcontroller that performs required calculations, measurements of the signal spectra and display of measurement results can be used as CD.

In order to increase the sensitivity of measurements of SLIF signals for subtracting fluctuations in the radiation intensity of the source, the corresponding PDDs can be made in the form of a pair of photosensitive elements (PSEs), for example, 5a and 5b (Picture 2, top view), which outputs are connected to various inputs of OAs 9. In this case, the in-phase signals of PSE are subtracted, while the SLIF signals are added to the random phases, as these phases are independent on each PSE, which leads to an increase in the resulting SLIF signal at the OA output 9 by approximately 1.4 times.

Since it is preferable to use different scattering angles for particles of different sizes, cuvette 2 and / or all of PDDs should have an ability of being moved along the axis of the radiation beam, i. e. with the possibility of changing the distance from cuvette 2 to PDDs 5, 6, 7. So that the range of the recorded angles' changes, but the relative position of PDDs remains constant and the conditions of processes in the tested system are kept unchanged.

As an illustration of the method and apparatus implementations, on Picture 3 there is experimental data on E. coli bacteria growth curves within the concentration range of 10 ³ -10 ¹⁰ CFU/ml, where 13 is the T signal change, 14 is the SLIF signal change (for one scattering angle, about 5 degrees), 15 is the resulting signal for one set of weights and degree of 2 for the T signal in the measured signal.

In case of a multiangle detection, the more the scattering angle is, the less the corresponding signal 15 is, while with the same morphology (particle size), the relative variations in the signals at different angles do not change over time This is schematically (solid curves) shown in Picture 4, but in case of aggregation of particles (aggregation into large aggregates) during the growth, the relative difference between manifested signals increases over time (dashed curves).

The registration growth curves of microorganisms in a liquid nutrient broth is a standard microbiological practice. For example, the bacterial growth occurs due to the division of bacteria and so increasing their number over time. If the bacteria do not aggregate, the turbidity of the suspension will also increase over time. Due to certain circumstances, for example, when antibacterial drugs are added to the liquid, bacteria may start to disintegrate thus decreasing the turbidity of the suspension. However, in case of the bacteria aggregation while the process of bacteria reproduction is still on, the turbidity of the suspension will decrease and this can be erroneously interpreted by both the SLIF signal and T signal as the destruction of bacteria. In this case, the appearance of relatively large bacterial aggregates lead to occurrence of low-frequency T fluctuations, which allow to distinguish the death of bacteria from their aggregation. On Picture 5 experimental data is shown an on the growth and "clumping" registration of Staphylococcus aureus bacteria, when bacterial aggregation leads to a false decrease in SLIF and T signals and simultaneous appearance of low-frequency fluctuations of the T signal, 16 - T signal; 17 - SLIF signal at one scattering angle.

When large contaminants pass through the radiation beam, the typical shape of the frequency spectrum of SLIF signals is distorted. Picture 6 shows the SLIF experimental frequency spectra at one scattering angle to measure the turbidity of a suspension of particles without contaminants (6a) and during the pass of a large foreign particle through a radiation beam (6b). The registration of the SLIF frequency spectrum allows to ignore the effect of large contaminants on the resulting signal.

Picture 7 shows SLIF signals at 4 different scattering angles (18-21, angles increase with the number increase) for the suspension of red blood cells. Cells settle and aggregate over time, while the dynamics of SLIF signals at different angles is different. Their ratio can be used for calibration by the size of aggregates of the studied cells.