WO1989000281A1

WO1989000281A1 - Transillumination imaging system using the antenna properties of heterodyne detection

Info

Publication number: WO1989000281A1
Application number: PCT/FR1988/000355
Authority: WO
Inventors: Daniel Dolfi; François Micheron
Original assignee: General Electric Cgr S.A.
Priority date: 1987-07-03
Filing date: 1988-07-01
Publication date: 1989-01-12
Also published as: FR2617601B1; FR2617601A1

Abstract

In a transillumination imaging system in which a body (H) is illuminated by transparency by a beam of light (FS, F1), a beam splitter (S1) derives from this beam (FS) a beam fraction (F'1) whose frequency is shifted by a frequency changer (CF). The beam (F3) obtained is combined with the signal beam (F2) produced by illumination of the medium (H). A detector (D) receives these beams and supplies a photocurrent having a frequency intermediate between the frequency of the signal beam (F2) and that of the frequency-shifted beam (F3). A privileged direction of the diffused light is thus isolated in the signal beam. Application: medical imaging.

Description

TRANS ILLUMINATION IMAGING SYSTEM USING HETERODYNE DETECTION ANTENNA PROPERTIES

The invention relates to an imaging system with visible wavelengths and in infrared in particular, applicable to any object embedded in a scattering medium. This system is particularly applicable to medical imaging and therefore relates to a transillumination or diaphanoscopy process.

The method described offers a completely harmless alternative to conventional medical imaging systems (X-rays associated with film recording, ultrasound, nuclear magnetic resonance, etc.). It must allow the formation of images, in two or three dimensions, by transmission of visible or near infrared light, through a living organ. The method can also be applied to any medium diffusing which is sufficiently transparent to light.

Transillumination as a diagnostic tool was used for the first time in 1929, for the detection of breast tumors. The article "Transillumination as an aid in the diagnosis of breast lesions" by M. CUTLER published in Surgery, Gynecology and Obstretics of June 1929, describes this means of diagnosis. The device used consisted only of a lamp placed under the patient's breast, the doctor directly observing the scattered light. All the devices used subsequently are improvements to this first system. The use of more powerful cooled torches (see the article DIaphanologie mammaire "by Ch. GROS et al published in the Journal de Radiologie - Tome 53 n ° 4 de 1972), sensitive surfaces in the near infrared (film and IR camera ), the formation of images at two different wavelengths, asso ciées to an important image processing, did not allow spectacular improvement. The main limitation comes from the very important value of the diffusion coefficient in biological tissues, even at wavelengths where absorption is low. Such devices are used only for mammography and do not allow the detection, with great safety, of small tumors (less than 1 cm), nor of deep tumors (even larger).

A known process, such as that carried out in the INSERM laboratory (U238) of Professor JARRY and described in the Doctoral thesis, specialty Sciences, defended by S .DEBRAY on June 12, 1987 before the University of Paris, Val de Marne and having for title "Device for the laser illumination trans of the biological tissues. Contributions of the temporal resolution and the Spectrophotometry", allows to escape the influence of the diffusion. The organ is lit by a laser beam. The detector, placed opposite the laser, behind the organ, is fitted with a collimator. The selection of flight times makes it possible not to take into account the scattered photons. Only the photons passed in a straight line are selected and a pulsed laser is used. This system also allows spectroscopic analysis of the organ using a dye laser. Contrast and perception of details are improved. Such a system is described in the article "Study in simulation of the behavior of light in biological tissues" by MAAREK et al published in Innovations des Techniques Biologiques et Médicales, Volume 7, n ° 3 de 1986.

The use of a slit scan camera (5 picoseconds of resolution) and a pulsed laser source, made it possible to develop a tomographic imaging system as described in the article "Laser Puise Tomography Using a Streak Camera by Y. TAKIGUCHI et al published in Proceedings "Image Detection and Quality" of July 86. Tomography is made possible by taking into account the backscattered signal. This technique, given the weakness of the signal, will be difficult to transpose to large living organs.

The low resolution of conventional systems, even when they are equipped with an efficient image processing, the complexity of the detection system of these methods leads us to propose a system making it possible to avoid these drawbacks.

The invention therefore relates to a transillumination system using the antenna properties of heterodyne detection. The very high angular selectivity of heterodyne detection, as well as the excellent signal-to-noise ratio that it allows, allow us to consider its application in the medical field.

The invention therefore relates to a transillumination imaging system comprising a laser source emitting a source light beam illuminating with a first beam a transparent or translucent medium as well as a detector receiving a second light beam resulting from the first beam after crossing said medium, characterized in that it comprises a beam splitter placed on the path of the source beam and diverting a portion of beam, a frequency changer receiving this portion of beam and providing in exchange a third beam slightly offset in frequency with respect to the first beam, an optical coupler placed on the path of the second beam also receiving the third beam and supplying the detector a fourth beam resulting from the combination of the second beam and the third beam.

The various objects and characteristics of the invention will appear more clearly in the description which follows, given by way of example with reference to the appended figures which represent:

FIG. 1, an exemplary embodiment of the imaging system according to the invention operating in two-dimensional imaging: - Figures 2 to 4, the operation of the system of the invention;

- Figure 5, an exemplary embodiment of the system according to the invention operating in three-dimensional imaging;

- Figure 6, the operation of the system of Figure

5;

- Figure 7, an embodiment of the system according to the invention illustrating a displacement exploration system;

- Figures 8 to 11, devices for performing an exploration of the body to Imager;

FIG. 12, circuits for processing the signals detected by the imaging system;

- Figures 13 to 21, different operating curves of the circuits of Figure 12; - Figure 22, an alternative embodiment of the circuits of Figure 12;

- Figure 23, another displacement system for exploring the body to be imaged;

- Figure 24, another device for exploring the system according to the invention;

- Figures 25 to 27, different operating variants of the detector of the system of the invention;

- Figure 28, a variant of the exploration system according to the invention; - Figure 29, a detailed view of the exploration device of Figure 28;

- Figure 30, a variant of the system according to the invention comprising several detectors to obtain an exploration of the body to be imaged; - Figure 31, a variant of the system exploration device according to the invention;

- Figure 32, a three-dimensional imaging system according to the invention comprising fiber optic transmission means; - Figure 33, a variant of the imaging system of Figure 32;

- Figure 34, a three-dimensional imaging system comprising cylindrical optical devices; - Figure 35, a three-dimensional imaging system according to the invention providing means for focusing on the detector;

- Figures 36 and 37, a three-dimensional imaging system according to the invention operating in pulses.

- Figure 38, an alternative embodiment of the imaging system according to the invention in which a frequency changer is located on the path of the light beam from the body to be imaged. Referring to Figure 1, we will first describe an embodiment of the system of the invention.

This system comprises a source L of radiation emitting a beam FS. This beam is, for example, a light beam of frequency ω A beam splitter SI, such as a semi-reflecting mirror, transmits part of the beam FS to a translucent or transparent body H. The latter receives the beam F1 and retransmits all or part of this beam in the form of a beam F2. This beam F2 is reflected by a mirror M2 towards a beam coupler S2. This beam coupler S2 can for example be a semi-reflecting plate.

The beam splitter SI reflects a part F'1 of the beam FS towards a mirror M1. This reflects the beam F'1 to a frequency changer CF which in exchange supplies a beam F3 of frequency ω + f. This beam F3 is transmitted to the beam coupler S2 which combines the beams F2 and F3, and supplies a beam F4 to a detector D. An optic C optionally focuses the light transmitted to the detector D.

In this way, the mixing is carried out on a detector of two optical waves F2 and F3 of slightly different frequencies. If the beam F3 has a frequency ω _S = _ω + f and the beam F2 has a frequency ω _L = _ω , we obtain a photocurrent I _SL at the intermediate frequency | ω _S -ω _L | :

^'

Detector area

(x, y) quantum efficiency of the detector and amplitude of the optical waves F2 and F3

respectively.

For a uniform detector and in the simplifying case of plane waves, we obtain, for two waves F2 and F3 non-collinear as shown in FIG. 2:

I _{SL α} SINC

The photocurrent therefore only has significant value when K _L and K _S are almost parallel with an angular tolerance: L

-

λ wavelength used

- ℓ characteristic dimension of the interaction zone If i _L and i _S are the photocurrents delivered when the detector receives waves of frequencies ω _L and ω _{S respectively} , we can consider in the simplifying hypothesis uniform and parallel plane waves:

The system of FIG. 1 transmits to the detector D two beams F2 and F3 in the form of a beam F4. This system makes it possible to use the antenna property of heterodyne detection. This property makes it possible to isolate a preferred direction in a diffuse signal, by means of a so-called local oscillator beam as shown in FIG. 3. In this figure, we find the beam F1, the body H, the beam F2, the beam F3, the semi-reflecting plate S2, the beam F4 and the detector D.

As shown in FIG. 4, the body H can be diffusing or include different zones. It receives the beam F1 and then retransmits a beam F2 which includes a beam F2 of non-scattered light and a part of scattered light represented by the light cone of FIG. 4.

At the wavelengths considered, the angular tolerance on the mixture of the two waves (F2 and F3) is of the order of 1 to 10 milliradlans. Under these conditions, practically only the part of the non-scattered light will be detected.

As soon as the local oscillator beam F3 is sufficiently intense, this collimation process has the following advantages: - greater luminous fluxes detected;

- very favorable signal-to-noise ratio thanks to the very low bandwidth required;

- same equivalent noise power as a perfect quantum detector limited by photon noise. The system of FIG. 1 thus makes it possible to produce a two-dimensional image of the body H. Thus heterodyne detection is used, in this case, as a means of collimation. It allows the measurement of the integrated collimated transmission on the crossing of the organ. The antenna properties make it possible to isolate, in the diffuse signal, the photons which have not been scattered and have followed a rectilinear path as shown in FIG. 4. The point-by-point analysis of the collimated transmission then makes it possible to reconstruct a two-dimensional image by projections of the entire organ on a plane orthogonal to the light beam. An image similar to that provided by conventional radiology systems is thus obtained.

The imaging system of the invention can also be produced to obtain a three-dimensional imaging of the body H. Figure 5 shows such a system. The organization of this system is similar to that of Figure 1.

The source L emits a light beam FS which is collimated by a lens B. In addition, an optic such as a lens O makes it possible to associate with a small volume of diffusing material (diameter 100 μm to 1 mm) a direction whose selection is performed by the local oscillator. We therefore measure the quantity of light scattered by a reduced volume of organ placed at the focal point object F of optics O. Any light scattered at another point is not collimated in the direction of the local oscillator. Thus a point by point analysis of the organ allows the reconstruction of a three-dimensional image. Indeed, translations of the lens O along two axes X and Y as shown in FIG. 6 allow point-to-point reconstruction of a plane PO located at a side Z of the axis Z. A translation of the lens O along the axis Z from Z _o to Z ₁ , then allows the formation of the two-dimensional image of the plane P1 of coast Z ₁ .

The operating principle of the system of FIG. 5 is identical to that of FIG. 1. However, it will be noted that the system of FIG. 1 is particularly sensitive to light which has not undergone scattering while the system of FIG. 5 makes it possible to detect the scattering points of the body H. Which amounts to saying that the image obtained by the system of FIG. 5 (in addition to being of three dimensions) is complementary to that obtained by the system of FIG. 1 .

In both systems, the source can transmit at multiple wavelengths; the simple installation of filters then allows spectroscopic knowledge of the medium and the formation of images at different wavelengths. It can be noted that the two-dimensional and three-dimensional images are formed by taking into account both the absorption and diffusion coefficients of the medium.

We will now describe the different elements of the invention and different embodiments.

The source L must be sufficiently spatially and temporally coherent to allow mixing on the detector. A laser source is therefore suitable. This source can be monochromatic, provide discrete wavelengths or a continuous spectrum (dye laser). Each wavelength then allows the formation of different images. The detector D, consisting of one or more photodiodes, for example, can be:

- single extended

- single point - linear multiple (bar of detectors)

- multiple matrix

Each type of detector is associated with one or more modes of lighting and scanning of the organ to be imaged.

When two beams of different sizes interfere with the detector, only the part of the detector, limited by the narrowest beam, generates a signal at the intermediate frequency. The angular selectivity and the transverse resolution are limited by the size of the narrowest beam.

We will therefore now describe various alternative embodiments of the system of the invention and of the device allowing exploration of the body H to be imaged. We will first describe various alternative embodiments of the two-dimensional imaging system of FIG. 1.

FIG. 7 represents a system in which an Image is obtained by translations along the axes T1 and T2 of the assembly E consisting of an optical device TR of a lens B of the beam coupler S2 and of the detector D.

The optical device TR is a device allowing the translation of the beam F3 along the axis T2. This device can for example be made with a set of two mirrors M3 and M4 parallel and movable in rotation as shown in Figures 8 and 9. We see in these figures that according to the orientation of the mirrors M3 AND M4, the beam F3 is translated in the direction T2. The optical device TR can also be produced, as shown in FIGS. 10 and 11, using prisms PR1, PR2, one of which is movable in the direction T2.

The device TR can therefore be produced using any system allowing translation of the beam F3 parallel to itself. In the device of FIG. 7, the translation of the beam F3 in the direction T2 is accompanied by an equivalent translation, of the frequency changer CF, of the lens B, of the coupler S2 and of the detector D so that the beam F3 scans the entire surface of detector D illuminated by the beam

F2. The wave surface of the beam F3 is of the same shape as that obtained at the output of the lens A (beam F2) throughout the translation T1.

According to the embodiment of Figure 7, the detector D is integral with the optical coupler S2.

The mirror M2 can be designed so as not to take into account all of the light leaving the body H. It reflects only a part of it, containing the collimated component to avoid dazzling the detector. Optical systems A and B can be dioptric or catodioptric. They supply from the source beams of the same wave surface, capable of interfering with the detector D.

The optical system A provides collimated lighting which covers the entire field to be examined.

The reflection devices M1 and M2 are either mirrors or prisms in total reflection. The optical coupler S2 is a semi-reflecting device which has reflection and transmission coefficients equal to 0.5. The beam splitter SI can have reflection and transmission coefficients such as 0, 2 for transmission and 0, 8 for reflection.

The light used is linearly polarized to obtain a mixture of the beams F2 and F3 without altering the contrast. In the system of FIG. 7, a polarizer P has been provided.

The frequency translation of the optical wave produced by the frequency translator CF is obtained either by an acousto-optical system, for example, a Bragg cell, or by mounting the mirror M1 on a piezoelectric material, the mirror M1, then driven by a voltage ramp, vibrates orthogonally to its plane. This frequency translation can also be generated by a half-wave plate in rotation between two quarter-wave plates.

The diameter of the detector is limited to that of the beam F2. Thus the sensitive noise generating surface is reduced to the maximum since it coincides with that generating the signal.

The transmission of the scattering medium is measured in a virtual direction, selected from the signal beam F2 extended by the local oscillator beam F3 of reduced dimensions. The amplitude of the signal at the intermediate frequency f is proportional (T _C = collimated transmission). This

allows natural compression of the dynamics of the processing electronics following the detector. The chain for detecting and processing the signal detected by the detector D is shown in FIG. 12. It is a conventional chain of which all the elements are well known. Let ω be the frequency of the optical source, therefore of the beam F2 and _ω + f the frequency of the local oscillator. This detection chain comprises a filtering circuit CD1 around the frequency f and amplification receiving the signal detected by the detector D. A rectifier CD2 is connected to the output of the circuit CD1. An integration circuit is connected to the rectifier CD2 and performs the integration of the detected signal. A low frequency filtering circuit CD3 thus provides a significant detected signal.

Figures 13 to 21 show the appearance of the signals at different points in the system.

Figures 13 and 14 represent, respectively, the evolution of the amplitude of the optical signal during a scan line, and the low frequency spectrum of this evolution.

The frequency f _C can be compared to a spatial frequency. It encrypts the transverse resolution of the system when scanning the scattering object. FIG. 15 represents the electromagnetic wave which constitutes the carrier of the signal obtained during a scanning line (electromagnetic wave at frequency _ω ).

Figure 16 gives the spectrum of this wave. It's the signal wave.

Figures 17 to 18 show the intermediate frequency signal f detected by the detector D and the associated frequency spectrum. This signal is obtained after mixing the signal beam and the local oscillator (output from CD1). Figures 19 to 20 show the rectified signal then filtered by the circuits of the detection chain (respectively at the output of CD2 and CD3).

FIG. 21 represents the frequency spectrum at the output of CD3. The assembly CD1, CD2, CD3 can be replaced by a filter followed by a synchronous detection, the reference of which is provided by the signal controlling the frequency changer.

FIG. 22 represents an alternative embodiment of the circuits of FIG. 12. This alternative comprises, after the filtering and amplification circuit CD1, a delay line CD5 and a logic gate CD6 placed in series between the circuit CD1 and the CD3 Integration circuit. This arrangement of circuits makes it possible, by intercorrelation, to further reduce the bandwidth of the electronics and to improve the signal-to-noise ratio of the system, the door CD6 is controlled by a signal of frequency f + δ f. The integration is made so as to provide a signal of frequency δ _f <<f. The amplitude of this signal always encrypts the collimated transmission of the diffusing medium. This improvement is applicable to all embodiments of the invention.

To reduce the response time of the system, it is possible to interpose between the coupler S2 and the detector an optical system not shown which allows the focusing of the beam F3 of local oscillator and of the signal beam F2 on a detector D of reduced size . The reduction of the sensi area ble is accompanied by the reduction in the response time of the detector, the coupler S2, this optical system and the detector D are integral during the translations T ₁ and T ₂ described above. The chain of detection circuits is that described above, with or without cross-correlation. There is always the possibility of focusing the beams F2 and F3 on the detector D.

The system of FIG. 23 gives rise to another variant embodiment, not shown, which differs from the previous one only by the fact that the coprs H is no longer displaced along the axes X, Y but it is the entire system. imagery that is moved. This is mounted on a U-shaped frame subjected to translations along the X and Y axes. The variant embodiment of the system of FIG. 24 is similar to that of FIG. 23 and its principle is the same apart from the translations of the body H. the beam F3 of the local oscillator and the lighting beam F2 have the same diameter, but they cover the entire field to be examined. A detector D whose sensitive surface has the size of the point to be resolved allows, by translations along the X and Y axes, the reconstruction of the image, the detector D can, in this case, be preceded by an optic which adapts the size of the virtual beam selected at the effective size of the detector D used as shown in FIG. 25.

Figures 25, 26 and 27 show different types of optics allowing either to focus a beam on detector D (figure 25) or to collimate it with a beam covering the whole detector (figure 26) or with a beam of more dimensions reduced (Figure 27).

In the variant embodiment of FIG. 28, the dimensions of the beam F3 and of the detector D are those of the area to be imaged. In this case, the detector D is stationary. The reconstruction of the image is obtained by translations along the X, Y axes of the signal beam F2. A TT system, allowing these translations, is a system with two parallel mirrors M6 and M7 rotating around two orthogonal axes as shown in Figure 29. The rotation around the axis Xa provides the translation along X, that around the the axis Ya provides the translation along Y. The mirror M2 can be large and remain stationary. It can be reduced in size but it must then move simultaneously with the mirrors M6 and M7 to follow the beam F2.

In FIG. 28, it can therefore be seen that the translation of the beam F1 using the TT device makes it possible to scan the body H to be imaged. The oscillator beam F3 has the dimensions of the area to be scanned and the detector D also. The scanning of the beam F1 therefore makes it possible to produce the image of the body H.

An alternative embodiment of the system of FIG. 28 provides for a detector not covering the entire area to be explored, the beam F3 of local oscillator still covers the entire field to be examined, but the detector has a surface of substantially the same dimensions. than the beam F2. This limits the transverse resolution and the detector D must move simultaneously with the beam F2.

It is obvious that in all the variants given above or below, it is always possible, after mixing the beams F3 and F2, to adapt their sizes to that of the detector by an optical system. FIG. 30 represents a variant providing a detection system D with several detectors. In this variant, a single detector, either point or extended, is no longer used, but a linear multiple detector of the photodiode array type, such a detector can be mounted in all the systems described above. It allows the suppression of a translation of the detector, or at least the reduction of its amplitude of translation.

The detector used is of the matrix type. If it is large enough to cover the entire field to be imaged, the image can be reconstructed by simple successive readings different point detectors. If the detector is not large enough, it suffices to associate it with the scanning systems described above. With an identical imaging field, such a detector allows the reduction of the scanning amplitudes.

The variant embodiment of FIG. 31 provides bundles F2 and F3 of the same dimensions. This variant provides for placing the beam translator TT upstream of the beam splitter SI. The beam translator TT controls the movement of the beam FS along the X and Y axes. There is thus no part in translation, apart from the detector D, thanks to the absence of any optical system between the separation and the beam recombination. The TT device described above, such as that of FIG. 29, allows translations along the X and Y axes. The four reflecting surfaces of the mirrors M6, M7 and M1, M2 maintain the coincidence of the beams F2 and F3 during the scanning of the body H. The movement of detector D is synchronous with that provided by the TT device. The mirror M1 is mounted on a piezoelectric material and performs the frequency displacement.

All the optical paths previously described take place in the air. They can also be done, after separation into beams F1 and F'1 in waveguides G1 and G2 such as optical fibers. This allows the TT device to be eliminated, the various translations being made by simple displacements of the waveguides G1 and G2. Figure 32 shows an example of assembly using waveguides. All the previous variants can be modified to accept optical fibers with input optics. Optical systems A and B then adapt to the output of these fibers. On the assembly described in figure 32: - the optical device A

- the M2 mirror

- the optical coupler S2

- optical device B

- the detector form an integral unit which is translated along the X and Y axes, so as to reconstruct the image. E ₁ and E ₂ are input optics in the single-mode fibers.

The diameters of the beams F2 and F3 are imposed by the transverse resolution which it is desired to achieve. The diameter of the detector if it does not have an optical adaptation determines this resolution.

For all 2D imaging systems. Collimated transmission measurements are performed point by point and stored in memory. They allow, after processing, the reconstruction of the image.

Referring to Figures 33 to 35, we will now describe alternative embodiments of the three-dimensional imaging system of Figure 5. As we saw in describing Figure 5, such a system allows the measurement of the intensity scattered at a point.

The system allows the measurement of the intensity scattered at a point of the scattering body H placed in coincidence with the object focus of the optical system O. The optical system O is an optical system with large digital aperture, that is to say a system ensuring a good collection of the diffused flux and a shallow depth of field of object. it ensures the shaping of the wave diffused at point F and therefore provides a plane wave which is superimposed in the coupler S2 with the beam F3 of the local oscillator. Any wave diffused elsewhere than at point F, does not fulfill at the output of the system O, the conditions of agreement with the beam F3 to supply an electrical signal at the frequency f. Only the light component which succeeds in leaving the medium without being re-emitted is taken into account. This advantage is very important for reconstructing a three-dimensional image.

The image detection and storage chain is identical to that described for the two-dimensional image.

Mirrors M1 and M2 are mirrors or prisms with total reflection. The separator S1 and the coupler S2 are semi-transparent plates or separator cubes. The device P is a polarizer.

The frequency translation is ensured in an identical manner to the above using the CF device.

One can provide as in the system of Figure 32, optical fibers G1 and G2 allowing transmission of the beams F1 and F'1. The devices E ₁ and E ₂ are then optical systems ensuring the coupling between the laser beam and the single-mode fibers.

The devices A and B are optical systems, at the fiber outlet, providing collimated beams.

All :

- optical devices A and B - optical system O

- M2 mirror

- S2 coupler

- detector D is integral and can be translated along the axes, X, Y and Z. It is thus possible to reconstruct different image planes in the object.

The beam from device A and the optical system O have the same axis.

The lighting beam F1 is reduced in size to obtain a fairly high power density at point F.

This system allows a mixture of F2 and F3 beams of large dimensions, since they have no direct relation to the required transverse resolution. In fact, the more the beams are extended, the smaller will be the diffusing volume taken into account at the point F.

For the exploration of the body H, provision has been made to represent in FIG. 33 a system having transmissions by waveguide. But it is possible without departing from the scope of the invention to apply to the three-dimensional imaging system the displacement exploration systems described above.

In particular, a variant of the system of FIG. 33 can consist in replacing one or both fibers by beam translation systems, of the type of those described above such as the TR or TT devices of FIGS. 7 and 28.

In this variant, the optical system A provides a collimated lighting beam sufficiently wide to cover the entire field to be imaged. In this case, it may not be translated, provided that the lighting is sufficiently uniform or that a prior measurement of the level of illumination or that the image processing allows for non-uniformity to be taken into account.

The light beam from device A may not be collimated but be focused, actually or virtually, inside or outside of the diffusing medium H.

The beams F2 and F3 from the optical devices O and B may not be collimated. waves F2 and F3 are not necessarily plane but they must in any case have the same geometric characteristics.

As shown in FIG. 34, the optical systems O and B may not be of revolution but be, for example, cylindrical. In this case, the detector D is then multiple and linear and arranged parallel to a generator of the cylindrical optics. The generators of the two optical systems O and B are then mutually parallel. The detector can be a strip of photodiodes parallel to these generators.

The device O shapes the wave coming from a line either by collimating it or by focusing it.

IF the array of detectors D covers the entire line studied, with good resolution, the device B is also cylindrical and generates a wave of the same shape as that supplied by the device O. If the bar is not large enough, the device O provides a collimated beam which the beam from the device B (which may be of revolution) comes to scan. This operation is analogous to the scanning described in relation to FIG. 7. The scanning is then accompanied by a translation of the detector.

The use of multiple linear or matrix detectors makes it possible, in the system of FIG. 33, to separate the detector D from the optical assembly made up of the optical devices A, O, B, C, of the mirror M2 and of the coupler S2, and thus to reduce, even to eliminate the translations of the detector D.

According to another alternative embodiment of the system of the invention, the selection of the planes to be imaged can also be done by the lighting beam. This variant is shown in Figure 35.

The semi-transparent plate S2 combines the two points F2 and F3. Only the component of the wave F2 coming from A which arrives outside the diffusing medium without having been diffused, interferes with the wave F3 of the local oscillator. It is thus seen that if the optical devices A, B, C, coupler S2 and detector D are integral, It is possible thanks to the translations along the axes X, Y, Z to reconstruct a three-dimensional image of the diffusing object .

The focuses at F2 and F3 can be done, in accordance with the variant embodiment of FIG. 34, along lines. The device C can then also be a cylindrical optical system and the linear detector D.

According to the invention, the foregoing systems can be the subject of an improvement which improves the collimation efficiency of the trans Illumination system, for this a pulsed operation of the system is adopted on emission and on detection. This variant requires frequency modulation of the signal beams F2 and local oscillator F3, which makes it possible to distinguish the non-scattered photons, not only angularly, but also according to their time of arrival on the detector D.

The laser source L provides long light pulses (for example from a few hundred ns to a few μs). The beam emitted is monochromatic. The frequency of the optical wave is ω. An acoustooptical device makes it possible to modulate the optical frequency within the pulses emitted. This frequency varies between the start of the pulse and its end between ω + f ₁ and ω + f ₂ .

Thus in FIG. 36, there is an acoustooptical device MAO1 placed on the path of the beam FS emitted by the source L.

A second acoustooptical modulator MAO2 is placed on the path of the beam F'1.

In the presence of the scattering medium H and the second acoustooptical MAO2 modulator, the optical paths are adjusted (for example by means of optical fibers) so that the light pulses arrive simultaneously on the detector.

The MAO2 acoustooptical device (or any other frequency translation device) translates the optical frequencies inside the pulse, by a value Δ f. The frequency of the pulses of the local oscillator therefore varies between ω + f ₁ - Δf and ω + f ₂ + Δ f.

The photoelectric detector is followed by a bandpass filter, not shown in the figure, of width δ f, centered around the intermediate frequency Δ f. If we consider the frequency diagram in the figure

37, at time t _i , (only the photocurrent at the frequency Δ f (to within δ f) can be detected (heterodyne detection). We know that the selectivity of heterodyne detection makes it possible to isolate the photons, originating from the scattering medium, and collinear with the local oscillator. Almost all of these photons have not been scattered, however photons having undergone several scatterings may have the good angular tuning characteristics. The Impulse operation, associated with the frequency modulation, makes it possible to evade their Influence. Indeed, when scattered photons, but satisfying the condition of angular agreement, reach the detector at time t. , they have a frequency of type _ω + f _i-1 . They therefore provide, by interfering with the local oscillator at ω + f _i-1 + Δ f, a photocurrent at the frequency | Δ f + f _i - f _i-1 | . The choice of δ f makes it possible to ignore such a signal.

The choice of δ f, τ (duration of the pulse), of f ₂ - f ₁ makes it possible to give the system a great temporal selectivity, adding to the angular selectivity.

A temporal selectivity δ t of the order of ten picoseconds makes it possible to distinguish photons having followed optical paths which differ by a value δ ~ 2mm (in water). During the Pulse, the optical frequency of the signal beam varies as _. : f ^f

We want that after detection and filtering, the component of the signal due to direct photons, of frequency f (t) and to scattered photons, angularly satisfactory, at frequency f (t- δ t), is not transmitted to the chain treatment. It is therefore necessary that:

It is possible to obtain a modulation f2-f1 of the order of 1 MHz during a pulse of duration 1 μs. This gives a temporal selectivity of 10 picoseconds for a bandwidth δ f of the order of 10 Hz. The frequency Δ f can be chosen independently, for example, of a few KHz.

In addition to the natural angular selectivity of heterodyne detection, there is a temporal selectivity which, in theory, makes it possible to completely dispense with diffusion. The choice of a very low bandwidth, upon detection, allows to improve the signal to noise ratio and to decrease the value of the equivalent noise power. (NEP = ν δ f).

If the frequency changer is of the acousto-optical type (Bragg cell), one can consider the frequency shift no longer on the local oscillator but on the signal beam as shown in Figure 38.

The angular selectivity of such a cell is thus used. Indeed the diffraction efficiency of such a cell varies. Indeed, the diffraction efficiency of such a cell varies greatly with the incidence of light on the acoustically generated network. Only the fraction of the scattered beam having an angle is diffracted and sees its

modified frequency. The angular tolerance on θ is of the order (n optical index of the modulator, Λ s

B e acoustic wavelength, e interaction length Δ θ _B , deviation at θ _B for which the diffraction efficiency is divided by 2).

As an example, we can have the following numerical values. v ~ 10 ³ ms ^-1 (wave speed in the crystal) f ~ 80.10 ⁶ Hz e ~ 5 mm

Δ θ B ~ 2 milliradians.

It is obvious that the above description has been given only by way of example. Other alternative embodiments can be envisaged without departing from the scope of the invention.

Numerical examples have been provided for illustrative purposes only to illustrate the description.

Claims

1. Transillumination Imaging System comprising a laser source (L) emitting a source light beam (FS) illuminating with a first beam (F1) a transparent or translucent body (H) and a detector (D) receiving a second light beam (F2) resulting from the first beam (F1) after crossing said medium (H), characterized in that it comprises a beam splitter (S1) placed on the path of the first source beam ( FS) and deriving a beam portion (F'1), a frequency changer (CF) receiving this beam portion (F'1) and providing in exchange a third beam slightly offset in frequency with respect to the first beam (F1) , an optical coupler (S2) placed on the path of the second beam also receiving the third beam (F3) and supplying the detector a fourth beam (F4) resulting from the combination of the second beam (F2) and the third beam (F3).

2. An imaging system according to claim 1, characterized in that the source (L) emits a beam (FS) of coherent light.

3. An imaging system according to claim 2, characterized in that the source (L) emits a beam (FS) of light which is linearly polarized.

4. An imaging system according to claim 1, characterized in that the coupler (S2) provides a fourth beam which is the combination of the second and third beams, placed in parallel.

5. An imaging system according to claim 1, characterized in that the second beam (F2) and the third beam (F3) have identical wave fronts.

6. Imaging system according to claim 1, characterized in that the second beam (F2) and the third beam (F3) are collimated plane waves.

7. Imaging system according to claim 1, characterized in that the frequency changer (CF) is an acousto-optical device.

8. An imaging system according to claim 1, characterized in that the frequency changer (CF) comprises a mirror (M1) mounted on an electrically excited piezoelectric crystal.

9. An imaging system according to claim 1, characterized in that the sections of the second and third beams are identical.

10. Imaging system according to claim 1, characterized in that the section of the third beam (F3) is less than that of the second beam (F2), the system further comprising a beam translator (TR) placed on the path of the third beam (F3) and making it possible to move its direction parallel to a fixed direction so as to explore on the detector (D) the area illuminated by the second beam (F2).

11. Imaging system according to claim 10, characterized in that the beam translator (TR) comprises two reflection surfaces (M3, M4) parallel and orientable around an axis located on the passage of the beam between the two mirrors M3 and M4.

12. An imaging system according to claim 10, characterized in that the beam translator (TR) has two parallel reflection surfaces (PR1, PR2) which can move away from one another.

13. An imaging system according to claim 9, characterized in that the body (H) is movable relative to the first beam (F1) and to the second beam (F2) along two axes (X, Y) located in a perpendicular plane to the direction of the first beam.

14. Imaging system according to claim 13, characterized in that the different optical devices transmitting the first beam, the second beam and the third beam are movable along the axis (X, Y) relative to the body (H) and that the detector (D) is fixed and of sufficient size to image an entire area of the body (H) to be imaged.

15. An imaging system according to claim 1, characterized in that the detector (D) is small in relation to the sections of the second and third beam and that it can move in a plane perpendicular to these beams.

16. Imaging system according to claim 1, characterized in that:

- The second beam (F2) is of smaller section than that of the third beam (F3);

- the surface of the detector (D) is equal to the section of the third beam;

- A beam translator is placed on the path of the first beam (F1) so as to move it parallel to a determined direction to allow exploration of the body (H).

17. Imaging system according to claim 9, characterized in that the first beam (F1) allows an exploration of an area of the body (H), the detector covering the section of the second and third beams and comprising a plurality of devices detection.

18. An imaging system according to claim 1, characterized in that it comprises a beam translator (TT) placed on the path of the source beam (FS) and making it possible to move the first and second beam (F2, F3) parallel to a given direction so as to explore an area of the body (H).

19. An imaging system according to claim 1, characterized in that it comprises a first optical guide (G1) receiving the first beam (F1) and transmitting it to the body (H), as well as a second optical guide (G2 ) receiving the third beam and transmitting it to the optical coupler S2.

20. An imaging system according to claim 19, characterized in that the first and second optical guide (G1, G2) are optical fibers.

21. Imaging system according to claim 1, characterized in that it comprises connected in series to an output of the electrical detection signal of the detector (D);

- a filtering and amplification circuit (CD1) centered on the difference of the frequencies of the second and third beam and receiving the signal detected by the detector (D); - a rectifier (CD2);

- an integrator (CD3);

- a low frequency filter (CD4).

22. Imaging system according to claim 21, characterized in that it comprises in series between, the filtering and amplification circuit (CD1) and the rectifier (CD2) a delay line (CD5) and a logic gate (CD6) controlled by a frequency signal slightly different from the frequency of the source beam (FS).

23. An imaging system according to claim 1, characterized in that it comprises on the path of the second beam an optical collimation device (O) whose focus (F) is located at a point on the body (H).

24. An imaging system according to claim 23, characterized in that the body (H) is movable relative to the first light beam (F1) and the second light beam (F2).

25. Imaging system according to claim 24, characterized in that it comprises optical guides (G1, G2) transmitting the first and the third beam (F1, F3) and making it possible to make the imaging system mobile relative to to the body (H).

26. Imaging system according to claim 23, characterized in that the optical collimation device (O) is a cylindrical lens and that the path of the third beam (F3) also comprises a cylindrical lens whose generatrix is parallel to that of the optical device (O), the detector comprising a row of detector devices aligned in a direction parallel to said generators.

27. An imaging system according to claim 23, characterized in that the detector (D) is preceded by a focusing lens (C) making it possible to project a point (f2) of the body (H) on the detector and to focus also the third beam (F3) on the detector (D);

28. Imaging system according to claim 1, characterized in that the source (L) provides light pulses and in that it comprises a first optical modulator (MAO1) placed on the path of the source beam and a second modulator (MAO2) placed on the path of said portion of the source beam (F'1).