Device for measuring radiation backscattered by a sample and measurement method using such a device

Abstract

A device for measuring radiation backscattered by a sample including: at least one light source that is configured to emit a light beam, along an axis of incidence, towards a surface of the sample so as to form, on said surface, an elementary illumination zone; an image sensor for forming an image of the radiation backscattered by the sample when the latter is illuminated by the light source, the image sensor lying in a detection plane; a bundle of optical fibres, extending, along an extension axis, between a proximal surface and a distal surface, the proximal surface being applied against the image sensor, the distal surface being configured to be applied against the surface of the sample;
wherein the light source is arranged around the bundle of optical fibres, and wherein the distance between the light source and the bundle of optical fibres is less than 1 mm.

Claims

1. A Device for measuring radiation backscattered by a sample including: at least one light source that is configured to emit a light beam, along an axis of incidence, towards a surface of the sample so as to form, on said surface, an elementary illumination zone; an image sensor for forming an image of the radiation backscattered by the sample when the latter is illuminated by the light source, the image sensor lying in a detection plane; a bundle of optical fibres, extending, along an extension axis, between a proximal surface and a distal surface, the proximal surface being applied against the image sensor, the distal surface being configured to be applied against the surface of the sample; wherein the light source is arranged around the bundle of optical fibres, and wherein the distance between the light source and the bundle of optical fibres is less than 1 mm.

2. The device according to claim 1, wherein the light source is arranged around the bundle of optical fibres, between the image sensor and the sample.

3. The device according to claim 1, wherein the light source is flush with the distal surface.

4. The device according to claim 1, wherein the extension axis is perpendicular to the detection plane.

5. The device according to claim 1, wherein the optical fibres of the bundle of optical fibres: are oriented in parallel to the extension axis; or form a bundle diverging or converging away from/toward the image sensor.

6. The device according to claim 1, including a plurality of light sources, each light source being arranged around the bundle of optical fibres.

7. The device according to claim 1, wherein at least two light sources emit light beams in different spectral bands, respectively.

8. The device according to claim 1, wherein at least one light source is such that the axis of incidence of the light beam produced by the latter intersects the extension axis of the bundle of optical fibres.

9. The device according to claim 1, wherein at least one light source is: a light-emitting diode; or a laser diode; or an end of an illuminating optical fibre, the illuminating optical fibre being coupled to an illuminating source.

10. The device according to claim 1, wherein the image sensor, the bundle of optical fibres and each light source are held securely to one another by means of a holding support.

11. The device according to claim 1, wherein the image sensor includes a matrix-array of pixels, such that the distance, parallel to the detection plane, between at least the light source and the matrix-array of pixels is less than 1 mm or less than 500 m.

12. A method for estimating a scattering and/or absorption optical property of a sample using a device according to claim 1, the method including: a) illuminating the sample using the light source of the device so as to form, on a surface of the sample, an elementary illumination zone; b) forming, using the image sensor, an image representing radiation backscattered by the sample under the effect of the illumination, the backscattered radiation emanating from the surface of the sample outside of the elementary illumination zone at several distances, referred to as backscattering distances, away from the elementary illumination zone; c) determining a scattering and/or absorption optical property according to an intensity of the backscattered radiation, at various backscattering distances, at least one backscattering distance being less than 1 mm or than 500 m.

Description

FIGURES

(1) FIGS. 1A and 1B show a device according to a configuration known from the prior art, termed the remote configuration.

(2) FIG. 1C schematically shows a device according to a contact configuration, of simple design.

(3) FIGS. 2A and 2B collate results of experimental measurements carried out on calibration samples, the optical properties of which are known, using devices according to the configurations presented in conjunction with FIGS. 1A and 1C. FIG. 2A shows a relative quantization error of an absorption coefficient according to the absorption coefficient of the calibration sample. FIG. 2B shows a relative quantization error of a reduced scattering coefficient according to the absorption coefficient of the calibration sample. FIG. 2C shows a variation in the reflectance of a tissue according to the backscattering distance for various values of the reduced scattering coefficient. FIG. 2D shows a variation in the reflectance of a tissue according to the backscattering distance for various values of the absorption coefficient.

(4) FIGS. 3A to 3G show various embodiments of a device according to the invention. FIG. 3H illustrates the main steps of a method for determining optical properties of the invention using a device such as shown in FIGS. 3A to 3C.

(5) FIGS. 4A and 4B collate results of experimental measurements carried out on calibration samples, the optical properties of which are known, using devices according to the configurations presented in conjunction with FIGS. 1A, 1C and 3A. FIG. 4A shows a relative quantization error of an absorption coefficient according to the absorption coefficient of the calibration sample. FIG. 4B shows a relative quantization error of a reduced scattering coefficient according to the absorption coefficient of the calibration sample.

DISCLOSURE OF PARTICULAR EMBODIMENTS

(6) The term optical scattering or absorption property is understood to mean an absorption coefficient, a scattering coefficient, or a reduced scattering coefficient, these quantities being known to a person skilled in the art and described in application EP3054281, the latter being included as a reference. On the basis of absorption properties, it is possible to determine a concentration of a component of the sample. For example, when the sample is the skin of a human or of an another animal, it is possible to determine a concentration of various components of the skin, as well as of the oxyhaemoglobin or of the deoxyhaemoglobin of the blood flowing through the skin. For example, considering the absorption coefficient .sub.a, it is possible to use an expression such that:
.sub.a=(.sub.i.sub.ic.sub.i) where c.sub.i is a concentration of components i in the skin (in mol.Math.L.sup.1), .sub.i is an extinction coefficient of each component i (in L.Math.mol.sup.1.Math.cm.sup.1); is a function relating c.sub.i and .sub.i to .sub.a.

(7) The term a or one is to be understood to mean at least one.

(8) FIG. 1A shows a configuration of the prior art, termed the remote configuration, according to which a light source 11 emits an illuminating beam 12 in a spectral band , so as to illuminate a sample 10 to be characterized. The sample lies along a surface S. The illuminating beam 12 passes through an aperture 121, and is then reflected towards the sample 10 by a reflecting element 122. It reaches the surface of the sample at an oblique incidence. The illuminating beam 12 forms, on the surface of the sample, a luminous trace, referred to as the elementary illumination zone 18. The elementary illumination zone corresponds to an intersection between the incident light beam 12 and the surface S of the sample. In the example shown, it is pointlike.

(9) Under the effect of the illumination, a portion of the light from the illuminating beam 12 scatters through the sample, and is backscattered thereby so as to form backscattered radiation 14 emanating from the surface S of the sample. The latter propagates towards an optical system 19 that is optically coupled to an image sensor 20. The surface of the sample then lies in an object focal plane of the optical system 19. Thus, the optical system 19 conveys the image of the backscattering radiation 14, emanating from the surface of the sample S, to the image sensor 20. Such a device is an application of the patent application EP3054281 cited in the prior art.

(10) The image sensor 20 may in particular be a CMOS or CCD sensor. It includes a carrier 21, including for example a readout circuit, on which a matrix-array of pixels 22 is arranged, confined within a protective enclosure 23. The matrix-array of pixels 22 lies in a detection plane P. The protective enclosure 23 includes a transparent window 24, lying parallel to the matrix-array of pixels 22. The term transparent is understood to mean transparent in the spectral band . Numerous electrical connections lie in the protective enclosure 23. In a plane parallel to the detection plane P, the protective enclosure extends over a thickness of 2 or 3 mm around the matrix-array of pixels 22.

(11) The path of photons scattering through the sample 10 to form the backscattered radiation 14 is schematically shown by dotted arrows. The backscattered radiation 14 emanates from the surface S of the sample 10 in various distinct elementary backscattering zones of the elementary illumination zone 18. The term backscattering distance refers to a distance separating an elementary backscattering zone from the elementary illumination zone 18. FIG. 1B is a view from above of the surface of the sample. FIGS. 1A and 1B show two elementary backscattering zones 13.sub.1 and 13.sub.2 lying at two backscattering distances d.sub.1 and d.sub.2, respectively, away from the elementary illumination zone 18, and from which emanate backscattered radiation 14.sub.1 and backscattered radiation 14.sub.2. The term elementary backscattering zone refers to a zone of the surface of the sample whose size is such that the zone may be considered to be located the same distance away from the elementary illumination zone 18. An elementary backscattering zone may be pointlike. The area of a backscattering zone is then smaller than 1 mm.sup.2, or even smaller than 100 m.sup.2. An elementary backscattering zone may also describe all or part of a thin ring, the thickness of which is preferably less than 1 mm, or even less than 500 m, or even 100 m or 50 m, extending around the elementary illumination zone 18.

(12) As mentioned in EP3054281, the more the backscattering distance decreases, the more the depth of the path of the photons forming the backscattered radiation through the sample decreases. Thus, the more the backscattering distance decreases, the more the backscattered radiation is representative of a surface portion of the sample. Conversely, the more the backscattering distance increases, the more the backscattered radiation is representative of a deep portion of the sample.

(13) By means of the optical system 19, each pixel of the matrix-array of pixels 22 is optically coupled to an elementary backscattering zone. It is understood that this device makes it possible to detect radiation backscattered over small backscattering distances, typically of a few tens of microns, thereby making it possible to correctly analyse the portion of the sample closest to the surface, extending a few tens of microns below the surface S. However, the device is bulky. It is not suitable for portable use, being worn by a human or another animal.

(14) The inventors have tested a substantially more compact configuration shown in FIG. 1C. It includes an image sensor 20, similar to the sensor described in conjunction with FIG. 1A. A point light source 11, for example a light-emitting diode, is arranged on the carrier of the image sensor 21. The image sensor 20 is applied directly against the sample 10. However, in such a configuration, the distance, parallel to the surface S of the sample, between the centre of the light source 11 and the matrix-array of pixels 22, may not be less than 3 or 4 mm. This is due to the dimensions of the protective enclosure 23 extending around the matrix-array of pixels 22. Because of this, in such a configuration, the backscattering distance cannot be less than a minimum distance d.sub.min, here equal to 4 mm, which makes determining the scattering and/or absorption optical properties of the portion of the sample closest to the surface thereof inconceivable. Moreover, even when the optical properties of the sample are uniform, such a distance proves to be too great for correctly characterizing the scattering and/or absorption properties of the sample, as explained below.

(15) The inventors have conducted experimental tests with a view to comparing the performance of the configurations shown in FIGS. 1A and 1C, respectively. To achieve this, they used calibration samples taking the form of phantoms, the scattering and absorption optical properties of which are known. For the tests, the following components were used: Light source 10 for the remote configuration of FIG. 1A: Power Technologies Inc. laser diode with wavelength =690 nm. Light source 10 for the contact configuration of FIG. 1C: Light-emitting diode (Kingbright KPTD-1608SURK) emitting in a spectral band centred on a wavelength of =640 nm and having an angle of emission of about 60. Image sensor 20: UI-1492-LE CMOS sensor by IDS Imaging.

(16) In the remote configuration, the optical system 19 is a Xenoplan 1.4/23-0902 objective23 mm by Schneider Kreuznach, arranged at a distance of 90 mm away from surface of the sample. An optical density, producing an attenuation of 10-2, was arranged between the laser and the mirror 122. Two polarizing filters are also arranged between the light source 11 and the sample 10 and between the sample 10 and the sensor 20, respectively. The optical density and the polarizing filters are optional and are not shown in FIG. 1A. The polarizing filters make it possible to prevent the diffuse reflection of the incident beam 12 off the surface of the sample reaching the image sensor 20.

(17) Various calibration samples were used, these having one and the same reduced scattering coefficient .sub.s and various absorption coefficients pa. During each test, the scattering and absorption coefficients were determined according to the method described in EP3054281, In general, the coefficients (.sub.a, .sub.s) are obtained by comparison between: a reflectance measured at various backscattering distances; a reflectance that is modelled, at each backscattering distance, by taking into account various reduced scattering coefficients .sub.s and various absorption coefficients .sub.a.

(18) The pair (.sub.a, .sub.s) is then determined, which pair best corresponds to the reflectance measured at the various backscattering distances.

(19) The diffuse reflectance corresponds to an intensity of detected backscattering radiation normalized by the intensity of the light beam incident on the sample, and may be assigned corrective factors, taking into account the response of the components used, for example the image sensor, the optical components or the light source.

(20) The remote configuration was first implemented, and the pair (.sub.a, .sub.s) was determined for six calibration samples on the basis of the backscattered signal at various backscattering distances, comprised between 500 m and 5 mm, according to a sampling interval of 80 m. The results of this test are represented by a dot in FIGS. 2A and 2B.

(21) FIG. 2A shows the relative error affecting the determination of the absorption coefficient .sub.a (ordinate axis) as a function of the actual absorption coefficient .sub.a of each sample (abscissa axis). FIG. 2B shows the relative error affecting the determination of the reduced scattering coefficient .sub.s (ordinate axis) as a function of the actual absorption coefficient .sub.a of each sample (abscissa axis).

(22) The contact configuration was implemented based on an assumption regarding the value of the reduced scattering coefficient .sub.s. The absorption coefficients .sub.a for each calibration sample were thus determined. The results of this test are represented by a triangle in FIG. 2A.

(23) The contact configuration was implemented without any assumption regarding the value of the reduced scattering coefficient .sub.s. The absorption and reduced scattering coefficients (.sub.a, .sub.s) for each calibration sample were thus determined. The results of this test are represented by a circle in FIGS. 2A and 2B.

(24) The following conclusions are drawn from the results presented in FIGS. 2A and 2B: the remote configuration provides optimal performance and may be considered a reference method; the contact configuration with an assumption regarding the reduced scattering coefficient provides an acceptable estimate for the absorption coefficient (cf. FIG. 2A), with an error range less than or of the order of 5%; the contact configuration without an assumption regarding the reduced scattering coefficient may result in a substantial measurement error, possibly reaching 15%, affecting the determination of the absorption (FIG. 2A) and reduced scattering (FIG. 2B) coefficients. Moreover, the errors respectively affecting these two coefficients appear to be correlated in terms of absolute value.

(25) The inventors have attributed the poorer performance of the contact configuration to the fact that when the minimum backscattering distance (d.sub.min) is small, typically less than 1 mm, the reflectance mainly varies according to the reduced scattering coefficient. FIGS. 2C and 2D illustrate this observation. These figures respectively show the variation in the reflectance (ordinate axis) according to the backscattering distance (abscissa axis) while considering various reduced scattering coefficients .sub.s and various absorption coefficients .sub.a, respectively. These reflectance values were obtained by modeling the scattering equation. In these two figures, the arrows represent increasing values of the coefficients .sub.a (in FIG. 2D) and .sub.s (in FIG. 2C).

(26) FIG. 2C was obtained by varying .sub.s from 10 cm.sup.1 to 80 cm.sup.1 by intervals of 4 cm.sup.1, the absorption coefficient .sub.a of the modelled sample being set at 1 cm.sup.1. The refractive index of the sample is set at 1.33, the sample being placed in air, the refractive index of which is equal to 1. The anisotropy coefficient g is set at 0.8. The anisotropy coefficient, known to a person skilled in the art, relates the reduced scattering coefficient .sub.s to the scattering coefficient .sub.s according to the expression:
s=.sub.s(1g)

(27) FIG. 2D was obtained by varying .sub.a from 0.01 cm.sup.1 to 4 cm.sup.1 by intervals of 0.2 cm.sup.1, the absorption coefficient .sub.s of the modelled sample being set at 20 cm.sup.1. The refractive index of the sample is set at 1.33, the sample being placed in air, the refractive index of which is equal to 1. The anisotropy coefficient g is set at 0.8.

(28) It is observed that when the backscattering distance is less than 1 mm, the reflectance hardly varies with .sub.a. Conversely, at small backscattering distances, the reflectance varies with .sub.s. Beyond a backscattering distance of 1 mm, the intensity of the backscattered radiation decreases both with .sub.a and with .sub.s. Considering such a variation, it is therefore difficult to estimate both the absorption coefficient .sub.a and the reduced scattering coefficient .sub.s. Specifically, in the absence of a short-distance measurement, i.e. a measurement at a backscattering distance of less than 1 mm, it is difficult to simultaneously estimate the absorption .sub.a and reduced scattering .sub.s coefficients, since the reflectance, i.e. the normalized intensity of the backscattered radiation, is affected in the same way by these two coefficients.

(29) Having observed the above, in order to be able to satisfactorily estimate the absorption and scattering properties using a compact and wearable device, the inventors have designed a device such as shown in FIG. 3A. The device includes an image sensor 20 such as described above. It includes a light source 11, capable of producing an illuminating beam 12 propagating towards the sample. The intersection of the illuminating beam 12 with the surface S of the sample forms an elementary illumination zone 18. The elementary illumination zone may be pointlike, i.e. having an area of less than 5 mm.sup.2, preferably less than 2 mm.sup.2 or 1 mm.sup.2. It may also form a pattern, the size of which is less than 5 mm, or even 2 mm or 1 mm. It may for example be a thin line. In FIG. 3A, the illuminating beam 12 is not shown because the device 1 is applied against the sample 10.

(30) The light source 11 may for example emit the illuminating beam 12 in a spectral emission band lying in the visible or near-infrared portion of the spectrum, i.e. between 400 nm and 950 nm. Preferably, the bandwidth is less than 100 nm, or even less than 50 nm or 20 nm. The term bandwidth is understood to mean the full width at half-maximum of the spectral emission band.

(31) In the example shown in FIG. 3A, the light source 11 is a light-emitting diode such as described in conjunction with the preceding examples. Alternatively, the light source may be a low-power laser diode. It may also be an end of an optical fibre, the optical fibre being coupled to a light source.

(32) The device includes a bundle of optical fibres 25, also referred to as a fibre-optic plate. The bundle of optical fibres extends between a proximal surface 25.sub.p, arranged against the image sensor 20, and a distal surface 25.sub.d, which is intended to be applied against the sample 10 to be analysed. The bundle of optical fibres 25 includes a plurality of optical fibres that are joined to one another. The optical fibres are bonded to one another. The bundle of optical fibres may include several hundred, or even several thousand, optical fibres. They allow an image of the backscattered radiation 14 emanating from the surface S of the sample to be conveyed towards the matrix-array of pixels 22 of the image sensor 20. In this example, the optical fibres of the bundle 25 lie along an axis Z, referred to as the axis of extension, such that the bundle 25 allows the image formed on the surface of the tissue to be conveyed, along the axis of extension, towards the matrix-array of pixels 22. The bundle of optical fibres is preferably joined to the image sensor. The optical fibres of the bundle lie in parallel along the axis of extension Z, as described in conjunction with FIGS. 3A, 3C, 3D and 3G, or may be inclined with respect thereto, as described in conjunction with FIGS. 3E and 3F. When they are inclined with respect to the axis of extension, they may form a bundle diverging or converging away from/toward the image sensor 20. The proximal surface 25.sub.p of the bundle lies parallel to the detection plane P, while the distal surface 25.sub.d of the bundle lies parallel to the surface S of the sample. Preferably, the detection plane is parallel to the surface of the sample. This is particularly the case when the examined sample is deformable, when, by applying the device against the sample, the surface of the sample S lies parallel to the distal surface 25.sub.d of the bundle. In this example, the proximal surface 25.sub.p is bonded to the transparent window 24. In this example, the axis of extension Z of the bundle 25 is orthogonal to the detection plane P, which corresponds to a preferred configuration.

(33) Through the use of the bundle 25 into all or part of the matrix-array of pixels 22, the optical fibres of the bundle of fibres open onto pixels that are adjacent to one another.

(34) Preferably, the area of the proximal surface 25.sub.p is smaller than or equal to the area of the matrix-array of pixels 22. Thus, in a transverse plane XY in which the proximal and distal surfaces lie, the beam 25 is inscribed within the perimeter defined by the matrix-array of pixels. This allows the light source 11 to be arranged against the bundle 25 so as to minimize the distance d.sub.min, parallel to the surface S of the sample (or in the detection plane P), between the light source 11 and the matrix-array of pixels 22. Thus, the device makes it possible to detect a backscattered signal emanating from the surface of the sample S over a minimum backscattering distance d.sub.min that is shorter than that of the configuration described in conjunction with FIG. 1C. This distance corresponds to the shortest distance, parallel to the plane of the sample, between the light source 11 and the bundle of optical fibres 25. Preferably, the distance d.sub.min is shorter than 1 mm, and even more preferably shorter than 500 m. It corresponds to a distance, parallel to the detection plane, between the axis of incidence of the beam produced by the light source and the matrix-array of pixels 22. Using the bundle 25 of optical fibres, the area of which is smaller than or equal to the area of the matrix-array of pixels 22, allows the distance, parallel to the surface of the sample, or in the detection plane P, between the light source 11 and the pixels of the image sensor 20 to be significantly decreased.

(35) It should be noted that using a bundle 25 of optical fibres allows the backscattered radiation of a sample to be analysed using a standard image sensor 20, without requiring the latter to be modified.

(36) Depending on the desired spatial resolution to be obtained, the diameter of each fibre of the bundle may be between 5 m and 100 m or 200 m, for example 70 m. In order to decrease leakages of light between adjacent fibres, fibres of the bundle may advantageously be surrounded by an opaque material.

(37) The light source 11 may be flush with the distal surface 25.sub.d of the bundle of fibres 25, or be arranged set back with respect thereto, for example by a distance of less than 1 cm, or even less than 5 mm. Bringing the light source 11 close to the distal surface 25 makes it possible to limit the area of the elementary illumination zone 18, due to the divergence of the incident light beam 12.

(38) The distal surface 25.sub.d of the bundle 25 may be coated with a transparent biocompatible material, facilitating use of the device on the skin of a human or of another animal.

(39) The inventors have used a bundle of optical fibres of 70 m in diameter, with a numerical aperture of 0.84, supplied by Paradigm Optics, reference PA0470, having a diameter of 1 inch in the plane XY and a thickness of inch along the axis Z. This has allowed a minimum backscattering distance d.sub.min equal to 470 m to be obtained.

(40) The image sensor 20, the light source 11 and the bundle 25 may be held securely by a holding support 26, the latter possibly being an encapsulation material made of rigid plastic.

(41) According to the examples shown in FIGS. 3A, 3B and 3C, the device includes a plurality of light sources 11. FIG. 3B shows various light sources 11.sub.1 . . . 11.sub.i . . . arranged around a bundle of optical fibres 25, at the smallest possible distance away therefrom, this distance being for example smaller than 1 cm or than 1 mm. Light sources emitting in spectral bands differing from one another may then be used. This allows the optical properties of the object to be characterized in the various spectral bands used. Preferably, the spectral bandwidth of each light source is less than 100 nm, less than 50 nm or less than 20 nm. The term bandwidth is understood to mean the full width at half-maximum of the emission peak corresponding to the spectral band.

(42) In FIGS. 3A and 3C, the optical fibres are parallel to the axis of extension Z.

(43) Each light source is preferably included within the holding support, such that the device forms a one-piece part.

(44) According to one embodiment, shown in FIG. 3C, the device includes a light source 11 emitting an incident beam 12 propagating along an axis of incidence that is inclined with respect to the axis of extension Z along which the incident bundle 25 lies. The angle of inclination between the axis of incidence and the axis of extension Z may for example be between 10 and 70 to 80. It is preferably between 30 and 45. This allows the elementary illumination zone 18 to be brought closer to the matrix-array of pixels 22 in the plane defined by the surface of the sample. The angle of inclination may for example be 35.

(45) According to one embodiment, shown in FIG. 3D, the device includes a first light source 11.sub.1 that is optically coupled to an auxiliary bundle 25 of optical fibres 25, extending between the first light source 11.sub.1 and a surface of the holding support 26 that is intended to be applied against the surface of the sample 10. Optionally, the device includes a second light source 11.sub.2 that is optically coupled to a second auxiliary bundle 25 of auxiliary optical fibres. The latter is configured to direct the light produced by the second light source 11.sub.2 along an axis of incidence forming an angle of inclination with the axis of extension Z. As described in conjunction with FIG. 3C, the angle of inclination may be between 10 and 70 or 80. It is preferably between 30 and 45.

(46) According to embodiments shown in FIGS. 3E and 3F, the fibres of the bundle of optical fibres 25 are not all oriented parallel to the axis of extension Z. These optical fibres form a bundle converging around the axis of extension Z. In FIG. 3E, the bundle 25 converges towards the surface of the sample. This allows the backscattering distances to be brought closer to one another. In FIG. 3F, the bundle 25 diverges away from the image sensor, allowing sensitivity to be increased, since several fibres may thus be coupled to one and the same pixel of the image sensor.

(47) According to one embodiment, shown in FIG. 3G, at least one light source is fibre based, in the sense that an illuminating optical fibre 11.sub.f extends from the light source towards the sample. A light source 11, included within the holding support 26, and from which an illuminating fibre 11.sub.f extends to a surface of the holding support that is capable of being applied against the sample 10, is shown. Another light source 11, included within the holding support 26, and from which an illuminating fibre 11.sub.f extends to a surface of the holding support that is capable of being applied against the sample 10, such that an illuminating beam is propagated along an axis of incidence forming an angle of inclination with respect to the axis of extension Z, is also shown. The angle of inclination is similar to that described in conjunction with FIG. 3C or 3D. The end of the illuminating fibre 11.sub.f then forms the light source 11. The use of a fibre allows the light source to be brought closer to the bundle of optical fibres 25.

(48) The device may be connected to a processor 30, including a memory 32, in which instructions making it possible to determine the absorption and/or scattering optical properties on the basis of the backscattered radiation detected, over various backscattering distances, by the image sensor 20 are stored. The method may include the following steps, illustrated in FIG. 3H:

(49) Step 100: applying the device against the sample, in contact therewith. Preferably, after application, the surface S of the sample lies parallel to the distal surface 25.sub.d of the bundle 25.

(50) Step 110: illuminating the sample using the light source or the light sources and detecting radiation 14 backscattered by the sample over at least two different and nonzero backscattering distances.

(51) Step 120: on the basis of the backscattered radiation detected by the image sensor, estimating, using the processor 30, scattering and/or absorption optical properties on the basis of algorithms. The algorithms used are for example based on a comparison between measured intensity values of the backscattered radiation and values modelled by considering various optical property values, as described in EP3054281. Advantage is then taken of the fact that at least one backscattering distance under consideration is shorter than 1 mm, or even shorter than 500 m, which makes it possible to better estimate the optical properties.

(52) The device shown in FIG. 3A was used with a single light source, corresponding to the light-emitting diode described in conjunction with the above tests. The device was used on calibration samples in a similar manner to that in the tests described in conjunction with FIGS. 2A and 2B.

(53) FIG. 4A shows the relative error affecting the determination of the absorption coefficient .sub.a (ordinate axis) as a function of the actual absorption coefficient .sub.a of each sample (abscissa axis). This figure shows: the results implementing the remote configuration such as shown in FIG. 1A, these results corresponding to the points in the form of dark dots; the results implementing the contact configuration such as shown in FIG. 1C, these results corresponding to the points in the form of circles; the results implementing the configuration according to the invention, such as shown in FIG. 3A, these results corresponding to the points in the form of asterisks.

(54) FIG. 4B shows the relative error affecting the determination of the reduced scattering coefficient .sub.s (ordinate axis) as a function of the actual absorption coefficient .sub.a of each sample (abscissa axis). The points shown in FIG. 4B correspond to those of FIG. 4A.

(55) It is observed that using the bundle of optical fibres 25 affords a significant increase in the performance of the contact configuration, the relative error associated with determining the scattering or absorption coefficients being less than 5%. Performance levels close to those afforded by the remote configuration are thus obtained.

(56) The invention could be used to study the absorption or scattering optical properties of objects, and in particular of biological tissues, for example the skin of a human or of another animal. The absorption or scattering optical properties may make it possible to quantify the presence of chromophore elements in tissues, or to carry out measurements for oximetry, which can be used for diagnosis or to follow up on the treatment of individuals. The invention may have applications in other industrial fields, for example in the agri-food industry.