METHOD FOR DETERMINING A FOCAL LENGTH OF A PARTICLE IN A MEDIUM

Abstract

A method for determining a focal length of a particle in a medium, wherein the method comprises: providing a sample; emitting a coherent light beam to irradiate the sample, wherein a first part of the light beam is scattered by the particle; recording an interference image; computing for a set of positions an electric field from the interference image; generating a representation comprising, for each of said positions, an intensity value of the coherent light beam calculated from the computed electric field; finding two positions, a first of which lies in the sample, a second of which lies in the beam direction behind or in front of said first position; and determining the focal length from the two found positions.

Description

TECHNICAL FIELD

The present disclosed subject matter relates to a method for determining a focal length of a transparent particle in a medium.

BACKGROUND

The background of this disclosed subject matter lies in the field of observing colloids such as emulsions, cell cul-tures and bacteria in a medium, for example to study contamina-tion of water by bacteria or bacterial onset of urinary tract infections. Moreover, plastic particles in ocean water, impuri-ties in liquid foods or pharmaceuticals, cells or their orga-nelles in bodily fluids, or the like could be observed. When a transparent particle is observed that focuses light by scatter-ing, its focal length provides valuable information: The focal length can be correlated with other properties of the particle such as its size, shape, refractive index and, thus, the dry mass, with the amount of water or saccharides in a bacterium, etc., or be used to distinguish classes and sub-classes of par-ticles in a medium.

Commonly used methods for determining a particle’s optical properties, e.g. its refractive index, are flow cytometry or phase contrast microscopy. Flow cytometry is well-suited to ex-tract statistical information about particles at high through-put. However, it suffers from a low precision in determining a single particle’s optical properties and fails in observing particles in a static medium. Phase contrast microscopy, on the other hand, provides a visualisation of individual particles at high precision but only at limited throughput and depth of fo-cus. Moreover, it requires a difficult and demanding interpre-tation of two-dimensional images to extract the refractive in-dex and, if desired, the focal length of the particle.

Another viable method is digital holographic microscopy. Therein, a sample containing said particle is irradiated with a coherent light beam to obtain an interference image. From the interference image, a three-dimensional model of the irradiat-ing light beam is computed by applying a reconstruction algo-rithm onto the interference image, e.g., a forward or back propagation or projection algorithm. In the three-dimensional model, the positions of the particle and of its focus, and the focal length as their difference can be determined.

However, experiments have shown that the three-dimensional model obtained by common digital holographic microscopy is of-ten qualitatively unsuited for reliably determining the focal length of particles. On the one hand, the intensity of the light beam at the focus generally exceeds the intensity at the particle by far such that the particle cannot be distinguished from its focus, especially in the case of a small focal length. On the other hand, an inevitable background signal induced by a portion of the light that is scattered in the medium, i.e., not by the particle but by other (usually smaller) bodies in the medium (e.g., by Mie-scattering), is recorded in the interfer-ence image. This background signal impairs a contrast between light scattered by the particle and light not scattered by the particle and, hence, deteriorates a determination of its posi-tion in the model. To still determine the particle position precisely a tedious and computationally demanding comparison of the reconstructed model with models of scattering, e.g., with Mie theory, would have to be performed.

BRIEF SUMMARY

It is an object of the disclosed subject matter to over-come the limitations of prior art and to provide a method for determining a focal length of a transparent particle in a medi-um in an efficient and precise manner.

This aim is achieved by a method for determining a focal length of a transparent particle in a medium, comprising:

• providing a sample of the medium containing said particle;
• emitting, with a light source, a coherent light beam to irradiate the sample, wherein a first part of the light beam is scattered by the particle to create a scattered light beam which has a focus in a beam direction behind or in front of the particle;
• recording, with a camera, an interference image of the scattered light beam and a second part of the light beam that has not been scattered by the particle;
• computing, with a processor, for each one of a set of po-sitions which are three-dimensionally distributed in a space including the particle and the focus, an electric field of the first part of the light beam from the interference image;
• generating, with the processor, a representation of the light beam covering said positions, and comprising, for each of said positions, an intensity value of the coherent light beam calculated from the computed electric field at that position;
• finding, with the processor, two positions, a first of which lies in the sample, a second of which lies in the beam direction behind or in front of said first position, and the intensity value for each of which is greater than the respec-tive intensity values for its nearby positions; and
• determining, with the processor, the focal length from the two found positions;
• wherein, in the step of generating, for each position, the intensity value is calculated according to
• $\begin{array}{l}I\left(r,z\right)=2\cdot I_1\left(r,z\right)\cdot \\ \left(1+\cos \left(\Theta +\arctan 2\left({\displaystyle \sum _k{C}_k\cdot \mathrm{Re}{\left(E_1\left(r,z\right)\right)}^k,{\displaystyle \sum _l{D}_l\cdot \mathrm{Im}{\left(E_1\left(r,z\right)\right)}^l}}\right)\right)\right)\end{array}$
• with

r, z    being coordinates of the position, wherein z is a coor-dinate in an opposite direction to the beam direction and r denotes a pair of coordinates in a plane perpen-dicular to said opposite direction,
I(r,z)  being the intensity value for the position,
I1(r,z) being an intensity of the first part of the light beam for the position,
Θ       being a phase shift parameter,
k, 1    being indices of summation,
Ck, D1  being expansion parameters,
E1(r,z) being the computed electric field for the position, and
arctan2 denoting the four quadrant inverse tangent function.

It shall be noted that the electric field of the light beam is generally complex-valued and, thus, can be described by its real and imaginary parts or by its amplitude and phase. Moreover, the light beam has an intensity, i.e., a power per unit area, which is proportional to the square of the amplitude of the electric field.

The first position corresponds to the particle position and the second position to the focus position such that the fo-cal length can be determined as their difference. Applicants have found out that said calculation of each intensity value allows to achieve a high contrast between light scattered by the particle and light not scattered by the particle and a re-liable distinction of the particle from its focus; conse-quently, said two positions can be precisely found. Thus, the focal length can be efficiently determined directly from the representation rendering a complicated comparison of the inten-sity or electric field of the light beam with a model of scat-tering dispensable.

Moreover, the phase shift and expansion parameters used for calculating each intensity value allow for precisely deter-mining the respective focal lengths of a large set of different particles and media. For example, any of the phase shift and expansion parameters can be predetermined, e.g., for a similar sample or measurement, or estimated on the basis of prior knowledge. For instance, the phase shift parameter may be set to Π to reduce a background signal. On the other hand, these parameters may be adapted to a particular situation, e.g., to precisely determine the focal length of a specific bacterium in water. To this end, in an embodiment, one or several of the phase shift and expansion parameters can be determined itera-tively to optimise the generated representation, e.g., by ap-plying an iterative optimisation algorithm known in the art such as a Quasi-Newton algorithm or the like, which may, e.g., terminate when the particle and focus position are converged.

In a favourable variant of this embodiment, at least one of the phase shift and expansion parameters is determined by repeating the step of generating, wherein said at least one pa-rameter is varied to increase a difference between higher in-tensity values and lower intensity values comprised by the rep-resentation. This optimisation reduces the influence of the background which is typically represented by lower intensity values and results in more pronounced higher intensity values. The difference can, e.g., be a difference between a sum of all higher intensity values and a sum of all lower intensity val-ues, a RMS (root mean square) contrast, etc. For example, em-ploying the latter yields a smaller number of more pronounced intensity value maxima which can be considered first in the step of finding said two positions.

In an alternative or additional variant, at least one of the phase shift and expansion parameters is determined by re-peating the steps of generating and finding, wherein said at least one parameter is varied to increase a difference between the intensity value for the first position and the respective intensity values for its nearby positions and/or to increase a difference between the intensity value for the second position and the respective intensity values for its nearby positions. This determination particularly emphasises the respective in-tensity value for the particle and/or focus position compared to the intensity values for respective nearby positions, such that a local contrast is enhanced.

According to another favourable variant, at least one of the phase shift and expansion parameters is determined by re-peating the steps of generating and finding, wherein said at least one parameter is varied to increase a ratio between the intensity value for the first position and the intensity value for the second position. This facilitates a distinction of the intensity value for the particle position from the respective intensity values for the focus position and its nearby posi-tions, which is especially suited for precisely determining small focal lengths.

In an embodiment at least one of the phase shift and ex-pansion parameters is determined by

• a simulation of the steps of providing, emitting, re-cording, computing and generating to obtain a simulated repre-sentation in the step of generating and, for each position cov-ered by the simulated representation, an intensity of a simu-lated light beam; and
• by repeating the simulation of the step of generating, wherein said at least one parameter is varied to reduce a dif-ference between the intensity values comprised by the simulated representation and the intensities of the simulated light beam at the respective positions.

The medium, additional scatterers in the medium, the par-ticle’s surrounding, its shape, size, orientation, refractive index, etc. can be separately simulated and, if necessary, ad-justed. Thus, said at least one of the phase shift and expan-sion parameters can easily be determined either for a particu-lar sample or for a more general case, e.g., by performing a simulation of a sample comprising several particles with dif-ferent local surroundings. Moreover, this embodiment can, i.a., be used to predetermine at least one of said parameters and/or to set at least one of said parameters to an initial value to be used for iteration according to one of the abovementioned variants.

In another favourable embodiment at least one of the phase shift and expansion parameters depends on the position in the representation. This allows to consider different local vicini-ties, i.e., nearby positions particularly of the particle position and the focus position, respectively, by predetermining or determining said at least one parameter differently when calcu-lating the intensities values for the respective positions. Thereby, an enhanced precision in determining both the particle and the focus positions and, thus, the focal length can be achieved.

To facilitate a distinction between light scattered by particles and light not scattered by particles, it is advanta-geous when in the step of finding, only those positions are considered, for which the intensity value is greater than a predetermined threshold. Apart from the easier distinction, this allows for a faster finding of the two positions due to a reduced amount of data to be considered. In this embodiment, the threshold may be predetermined as known to the skilled per-son, e.g., as a percentage of a (general, typical, expected, or currently determined) maximum intensity value, as an average of intensity values for several positions, as an overall average of all intensity values of the representation, etc.

BRIEF DESCRIPTION OF THE DRAWINGS

The disclosed subject matter shall now be explained in more detail below on the basis of exemplary embodiments thereof with reference to the accompanying drawings, in which:

FIG. 1 shows a schematic side view of an inline interfer-ometer used for a method according to the present disclosed subject matter;

FIG. 2 shows a determination of a focal length according to the disclosed subject matter from an interference image rec-orded by the interferometer of FIG. 1 in a flow chart; and

FIG. 3 shows intensity values of a light beam determined according to the method illustrated in FIGS. 1 and 2 as a graph over an opposite direction to a direction of the light beam.

DETAILED DESCRIPTION

FIG. 1 shows an inline interferometer 1 comprising a light source 2 and a camera 3. The inline interferometer 1 is used to determine a focal length d of a (microscopic) transparent par-ticle P in a medium M. For determining the focal length d, a sample 4 of the medium M containing the particle P (or several particles P) is provided in a light path 5 between the light source 2 and the camera 3. Each particle P could be a cell, a bacterium, a charged particle, a micro plastic particle, an or-ganelle, etc., and the medium M could be water, oil, a body fluid (e.g. blood), a solution, or the like. Generally, the fo-cal length d of any transparent particle P focusing light in a medium M may be determined.

To this end, the light source 2 emits a coherent light beam 6 to irradiate the provided sample 4. The light source 2 may be of any type that is capable of emitting a coherent light beam 6, for example a laser diode.

In the sample 4, a first part of the light beam 6 is scat-tered by the particle P and, thereby, creates a scattered light beam 7 which has a focus 8 that is, in the example of FIG. 1, in a beam direction 9 behind the particle P, namely at a dis-tance therefrom corresponding to the focal length d behind the particle P. In other cases, where the particle P acts as a di-verging lens, the focus is in front of the particle P when seen in beam direction 9 (not shown).

However, a second part of the light beam 6 is not scat-tered by the particle P and traverses the sample 4 as an un-scattered light beam 10. In the context of the present descrip-tion, the scattered light beam 7 relates to scattering by the particle P in the medium M whereas the unscattered light beam 10 has not been scattered by the particle P in the medium M.

The scattered light beam 7 and the unscattered light beam 10 interfere with each other. At the end of the light path 5, the camera 3 records an interference image 11 (FIG. 2) of the scattered light beam 7 and the unscattered light beam 10. The camera 3 can, for the present purpose, be any analogue or digi-tal camera, e.g., with a Complementary Metal-Oxide Semiconduc-tor (CMOS) or a Charge Coupled Device (CCD) image sensor.

It shall be noted, that a portion of both said scattered light beam 7 and said unscattered light beam 10 may, however, also be scattered in the medium M, e.g., by other (usually smaller) bodies (e.g., via the Tyndall Effect), and that this portion of the unscattered light beam 10 slightly deviates from the beam direction 9. In the interference image 11, said por-tions of the scattered and unscattered light beams 7, 10 result in an inevitable, though undesirable background signal that im-pairs the determination of the focal length d.

The interferometer 1 may comprise one or more further op-tical devices as known in the art of holographic microscopy, e.g., an attenuator ring to improve the signal to noise ratio, a microscope objective, a phase plate, one or more lenses, or the like. Moreover, the interferometer 1 can also be embodied as a different type of interferometer than an inline interfer-ometer, for example as an interferometer 1 utilizing beam-splitters.

The interference image 11 recorded by the camera 3 is then forwarded to a processor 12 via an interface 13. Optionally, the processor 12 can preprocess the interference image 11 as known in the art, e.g., to numerically correct for aberrations. Subsequently, the processor 12 processes the interference image 11 to determine the focal length d therefrom as shall now be explained with reference to FIGS. 2 and 3.

While the camera 3 records the interference image 11 as a purely two-dimensional image, this interference image 11 en-codes both intensity and phase information of the light beam 6. This information allows the processor 12 to three-dimensionally “reconstruct” the light beam 6.

In a first step S₁, the processor 12 computes an electric field E₁ of the first part of the light beam 6, i.e., of the scattered light beam 7, for each one of a set of positions H from the interference image 11. The positions H are three-dimensionally distributed in a space R which includes the par-ticle P and the focus 8. In the examples of FIGS. 1 to 3, the space R ranges from within the sample 4 to the camera 3 as the focus 8 is known or expected to be behind the particle P (in beam direction 9). In another embodiment, the space R ranges from within the sample 4 to the light source 2 (or even be-yond), when the focus 8 is known or expected to be in front of the particle P (in beam direction 9). In yet another embodi-ment, the space R ranges, e.g., from the camera 3 to the light source 2 (or beyond).

In the example of FIG. 2, the positions H lie in several virtual planes 14₁, 14₂, ..., generally 14_i, which are perpendicu-lar to the beam direction 9 of the unscattered light beam 10. In this example, the positions H are identically arranged in each virtual plane 14_i; however, this is optional. In other ex-amples, the positions H may be distributed differently in the space R, e.g., in a regular grid arrangement representing ad-joining cubic regions or in an irregular grid arrangement rep-resenting adjacent arbitrarily formed spatial regions.

To compute the electric field E₁ of the scattered light beam 7, for each position H in the set, the processor 12 ap-plies a reconstruction algorithm onto the interference image 11. The electric field E₁ of the scattered light beam 7 for each position H is complex-valued and, thus, its real and imag-inary parts or its phase φ₁ and amplitude, respectively, are computed. In the present example, the processor 12 applies the reconstruction algorithm plane by plane in a direction 15 which is opposite to the beam direction 9. However, this is optional. Thus, multiple variants of reconstruction algorithms may be ap-plied as known in the art, for example forward or back propaga-tion or projection algorithms, e.g., an inverse Radon transfor-mation, a Fourier-domain reconstruction algorithm, an iterative reconstruction algorithm, etc.

In a subsequent second step S₂, the processor 12 generates a representation 16 of the sample 4 which covers all positions H and comprises, for each position H, a respective intensity value I (FIG. 3) which represents an intensity of the light beam 6 at that position H. From the electric field E₁ and an intensity I₁ of the scattered light beam 7 computed for that position H, e.g., as a square of the amplitude of the electric field E₁, the intensity value I for each position H is calcu-lated according to

$\begin{array}{l}I\left(r,z\right)=2\cdot I_1\left(r,z\right)\cdot \\ \left(1+\cos \left(\Theta +\arctan 2\left({\displaystyle \sum _{k\in K}{C}_k\cdot \mathrm{Re}{\left(E_1\left(r,z\right)\right)}^k,{\displaystyle \sum _{l\in L}{D}_l\cdot \mathrm{Im}{\left(E_1\left(r,z\right)\right)}^l}}\right)\right)\right)\end{array}$

with

• r, z being coordinates of the position H, wherein z is a co-ordinate in an opposite direction 15 to the beam direc-tion 9 and r denotes a pair of coordinates in a plane x, y perpendicular to said opposite direction 15,
• I(r,z) being the intensity value for the position H,
• I₁(r,z) being the intensity of the first part of the light beam 6 for the position H,
• Θ being a phase shift parameter which has either been predetermined or is determined as will be shown below,
• k, 1 being indices of summation which are iterated over a set K, L of values, respectively,
• C_k, D₁ being expansion parameters, which have either been pre-determined or are determined as will shown below,
• E₁(r,z) being the computed electric field for the position H, and
• arctan2 denoting the four quadrant inverse tangent function.

It shall be understood that the summations exclude the trivial cases of all expansion parameters C_k, D₁ being zero and of arctan2 solely yielding the phase φ₁ of the electric field E₁ of the scattered light beam 7. Moreover, the indices of sum-mation k, 1 do not have to be restricted to integer numbers, i.e., each of the sets K, L could, e.g., include real valued numbers as well.

In a third step S₃ subsequent to step S₂, the processor 12 finds two positions H_P, H_F: a first position H_P which is the particle position and lies in the sample 4, and a second posi-tion H_F which is the focus position and lies behind (or in oth-er cases: in front of) the particle position H_P when seen in beam direction 9, i.e., either within the sample 4 or between the sample 4 and the camera 3 (in said other cases: either within the sample 4 or between the sample 4 and the light source 2 or even beyond). The two positions H_P and H_F are found according to the following criterion: For each of these two po-sitions H_P, H_F the respective intensity value I_P, I_F is a local maximum, i.e., each intensity value I_P, I_F is greater than the respective intensity values I_NP, I_NF for nearby positions H_NP, H_NF (FIG. 3). The nearby positions H_NP are the positions H lo-cated in a vicinity, e.g., a sphere or box, around the particle position H_P. This applies correspondingly to the nearby posi-tions H_NF of the focus position H_F.

To find the two positions H_P, H_F the processor 12 may ap-ply any finding algorithm; e.g., it may first determine all in-tensity values I which show a local maximum and then compare the respective positions H in order to find the two positions H_P, H_F which are distanced from each other in beam direction 9. If several pairs of first and second positions H_P, H_F fulfil this selection criterion, either several particles P are pre-sent in the sample 4 (FIG. 1); otherwise, an additional selec-tion criterion may optionally be used, e.g., taking the posi-tion H with the highest intensity value I as the focus position H_F, or taking the position H with a distribution of intensity values I therearound showing a known particle scattering dis-tribution (e.g., a distribution derived from Mie scattering theory) as the particle position H_P, etc.

In a final step S₄, the processor 12 determines the focal length d of the particle P in the medium M from the two posi-tions H_P, H_F found in step S₃, e.g., as their mutual distance calculated from their respective coordinates x, y, z in a given coordinate system 18 (FIG. 3).

Various embodiments of calculating the intensity values I in step S₂ shall now be explained with respect to FIG. 3.

The exemplary graph of FIG. 3 shows intensity values I which have been calculated along the direction 15 and which are represented by a solid line 17. In a first embodiment, at least one of the phase shift and expansion parameters Θ, C_k, D₁, i.e., the phase shift parameter Θ, one or more of the first ex-pansion parameters C_k and/or one or more of the second expan-sion parameters D₁ is predetermined, e.g., for a similar sample or measurement, or determined on the basis of prior knowledge. For example, the phase shift parameter Θ may be set to Π, which generally reduces the background signal.

In an alternative or additional embodiment, at least one of the phase shift and expansion parameters Θ, C_k, D₁ is itera-tively determined for the recorded interference image 11 of the sample 4.

In a first variant thereof, step S₂ is iteratively repeat-ed (dashed branch b₁ in FIG. 2) and said at least one parameter Θ, C_k, D₁ is varied in each iteration to increase a difference between higher intensity values I and lower intensity values I comprised by the representation 16. Said higher intensity val-ues I may, for example, be higher than an average intensity value I_avg, higher than an intensity threshold I_th, e.g., a cer-tain percentage of a highest intensity value I (here: I_F), or may be a fixed number of (e.g.: ten) local maxima of the calcu-lated intensity values I comprised by the representation 16, etc. Similarly, said lower intensity values I may be lower than the abovementioned (or a different) intensity threshold I_th, said average intensity value I_avg, etc.

In this example, the difference between higher and lower intensity values I is computed as a sum of absolute deviations ΔI_avg between each intensity value I and the average intensity value I_avg, such that the variance of all intensity values I comprised by the representation 16 is increased. Alternatively, other differences may be employed, e.g., a difference between a sum of all higher and a sum of all lower intensity values I, etc.

In the example of FIG. 3, the result of the iterative de-termination of said at least one parameter Θ, C_k, D₁ is symbol-ised by the solid line 17, whereas a dashed line 17′ symbolises initial intensity values I′, i.e., intensity values calculated prior to repeating step S₂. As shown in this example, the solid line 17 has more pronounced maxima I_P, I_F than the dashed line 17′ which, besides, shows maxima in the intensity values I′_P, I′_F at different particle and focus positions H′_P, H′_F.

It shall be noted, that FIG. 3 depicts intensity values I in direction 15, i.e., in z-direction, while the variation of said at least one parameter Θ, C_k, D₁ generally affects the representation 16 in all three dimensions. It is further noted that the determination of the expansion parameters C_k, D₁ in-cludes a determination of the respective set K, L, i.e., the sets K, L may be predetermined and the expansion parameters C_k, D₁ be iteratively determined, or the sets K, L may be deter-mined iteratively as well.

In a second variant, the steps S₂ and S₃ are repeated (dashed branch b₂ in FIG. 2), and said at least one parameter Θ, C_k, D₁ is varied to increase a difference ΔI_P-NP between the intensity value I_p for the particle position H_P and the respec-tive intensity values I_NP for its nearby positions H_NP, wherein the intensity values I_NP of said nearby positions H_NP are op-tionally averaged to an average intensity value I_NP,_avg. Alterna-tively, a difference between the intensity value I_P and any other combination of nearby intensity values I_NP may be em-ployed, e.g., a weighted sum of the intensity values I_NP. Simi-larly, said at least one parameter Θ, C_k, D₁ is additionally or alternatively varied to increase a difference ΔI_F-NF between the intensity value I_F for the focus position H_F and the respective intensity values I_NF for nearby positions H_NF of the focus posi-tion H_F, e.g., their average I_NF,_avg.

In a third variant, the steps S₂ and S₃ are repeated, wherein said at least one parameter Θ, C_k, D₁ is varied to in-crease a ratio between the intensity value I_P for the particle position H_P and the intensity value I_F for the focus position H_F.

In a further additional or alternative embodiment, the processor 12 simulates a light beam irradiating a sample, its scattering and its intensities recorded in an interference im-age. This simulation may be performed as known in the art, e.g., using a field tracing method, either for an assumed par-ticular sample or for a more general sample, each sample con-taining one or more particles. From this simulated interference image, the intensity values I are calculated for a further set of positions, which are different or equal to the set of posi-tions H, by performing the steps S₁ and S₂ as illustrated above to obtain a simulated representation. Thus, the steps of providing, recording, emitting, computing and generating are simulated. Then, a repeated simulation of step S₂ of generating is performed, wherein said at least one parameter Θ, C_k, D₁ is varied to reduce a difference between the intensity values com-prised by the simulated representation and the intensities of the simulated light beam at the respective positions. Of course, the simulations may optionally be performed for differ-ent particles in different media, and the resulting parameters Θ, C_k, D₁ may be combined, e.g., averaged, to be applicable to a wide variety of particles P and media M.

Optionally, at least one of the phase shift and expansion parameters Θ, C_k, D₁ depends on the position H in the represen-tation 16 such that the intensity values I for different posi-tions H are calculated using different values of said at least one parameter Θ, C_k, D₁. For example, a value of said at least one parameter Θ, C_k, D₁, e.g., C₂, may be predetermined or de-termined for the particle position H_P and its nearby positions H_NP to be different from the value of the same at least one pa-rameter Θ, C_k, D₁ for the focus position H_F and its nearby posi-tions H_NF.

Any of the abovementioned embodiments and variants may be performed using optimisation algorithms with stopping criteria known in the art, e.g., a Quasi-Newton algorithm which stops when the particle and focus positions H_P, H_F are converged. Moreover, the abovementioned embodiments and variants may be combined while optimising a different one or different ones of said parameters Θ, C_k, D₁, respectively.

In an optional embodiment, the intensity threshold I_th is introduced prior to the processor’s step S₃. Thereupon, the po-sitions H for which the determined intensity value I is smaller than the calculated intensity threshold I_th shall not be con-sidered and used in step S₃. The intensity threshold I_th may, e.g., be estimated, or be a percentage of a maximum intensity value I (I_F in FIG. 3), a running average, an average of two or more, or even of all intensity values I comprised in the repre-sentation 16.

In a further optional embodiment, a reference image 19 of the coherent light beam 6 which has not been scattered by the particle P, thus relating to the unscattered light beam 10, is generated. Based thereupon, the processor 12 may normalise the interference image 11 prior to said step S₁ of computing as known in the art, such that a normalised computed electric field E₁ is computed in step S₁ and used as the electric field E₁ in subsequent step S₂.

It goes without saying, that the present method may be performed using multiple light beams at different frequencies and/or multiple light sources, e.g., to study the focal length from different illuminating angles and/or a dispersion of the particle P. Moreover, in addition to the focal length d, the intensities I_P, I_F at the particle and focus positions H_P, H_F may be used as well, e.g., to characterise or discriminate the particle P from other particles in the sample 4.

The disclosed subject matter is not restricted to the spe-cific embodiments described above but encompasses all variants, modifications and combinations thereof that fall within the scope of the appended claims.

Claims

1. A method for determining a focal length of a transparent particle in a medium, comprising: providing a sample of the medium containing said particle;emitting, with a light source, a coherent light beam to irradiate the sample, wherein a first part of the light beam is scattered by the particle to create a scattered light beam which has a focus in a beam direction behind or in front of the particlerecording, with a camera, an interference image of the scattered light beam and a second part of the light beam that has not been scattered by the particle;computing, with a processor, for each one of a set of positions which are three-dimensionally distributed in a space including the particle and the focus, an electric field of the first part of the light beam from the interference image;generating, with the processor, a representation of the light beam covering said positions, and comprising, for each of said positions, an intensity value of the coherent light beam calculated from the computed electric field at that position ;finding, with the processor, two positions, a first of which lies in the sample, a second of which lies in the beam direction behind or in front of said first position, and the intensity value for each of which positions is greater than the respective intensity values for nearby positions; anddetermining, with the processor, the focal length from the two found positions;wherein, in the step of generating, for each position, the intensity value is calculated according to Ir,z=2⋅I1r,z⋅1+cosΘ+arctan2∑kCk⋅ReE1r,zk,∑lDl⋅ImE1r,zlwithr, z being coordinates of the position, wherein z is a coordinate in an opposite direction to the beam direction and r denotes a pair of coordinates in a plane x, y perpendicular to said opposite direction,I(r,z) being the intensity value for the position,Il(r,z) being an intensity of the first part of the light beam for the position,Θ being a phase shift parameter,k, 1 being indices of summation,Ck, D1 being expansion parameters,E1(r,z) being the computed electric field for the position, andarctan2 denoting a four quadrant inverse tangent function.

2. The method according to claim 1, wherein at least one of the phase shift and/or expansion parameters is determined by repeating the step of generating, wherein said at least one of the parameters is varied to increase a difference between higher intensity values and lower intensity values comprised by the representation.

3. The method according to claim 1, wherein at least one of the phase shift and/or expansion parameters is determined by repeating the steps of generating and finding, wherein said at least one of the parameters is varied to increase a difference between the intensity value for the first position and the respective intensity values for nearby positions and/or to increase a difference between the intensity value for the second position and the respective intensity values for nearby positions.

4. The method according to claims 1, wherein at least one of the phase shift and/or expansion parameters is determined by repeating the steps of generating and finding, wherein said at least one of the parameters is varied to increase a ratio between the intensity value for the first position and the intensity value for the second position.

5. The method according to claims 1, wherein at least one of the phase shift and/or expansion parameters is determined by a simulation of the steps of providing, emitting, recording, computing and generating to obtain a simulated representation in the step of generating and, for each position covered by the simulated representation, an intensity of a simulated light beam; andby repeating the simulation of the step of generating, wherein said at least one of the parameters is varied to reduce a difference between the intensity values comprised by the simulated representation and the intensities of the simulated light beam at the respective positions.

6. The method according to claims 1, wherein at least one of the phase shift and/or expansion parameters depends on the position in the representation.

7. The method according to claim 1, wherein, in the step of finding, only those positions are considered, for which the intensity value is greater than a predetermined threshold.