WO2022200554A1

WO2022200554A1 - Optical system for reading radiochromic films and corresponding method of operation

Info

Publication number: WO2022200554A1
Application number: PCT/EP2022/057892
Authority: WO
Inventors: Julien JURCZAK; Benjamin RAPP; Robert Simcoe
Original assignee: Commissariat A L'energie Atomique Et Aux Energies Alternatives
Priority date: 2021-03-25
Filing date: 2022-03-25
Publication date: 2022-09-29
Also published as: FR3121233B1; FR3121233A1

Abstract

The invention relates to an optical system (1) for reading a radiochromic film, comprising: - a light source (2) comprising at least one light-emitting diode (LED) arranged to illuminate the radiochromic film, the emission wavelength of the LED being suitable for matching that of the maximum absorption peak of the radiochromic film, - a bi-telecentric lens (3) arranged to receive the optical beam which is emitted by the diode(s) (LED) and has passed through the radiochromic film, - an imager (4) optically coupled to the bi-telecentric lens.

Description

Title: Optical system for reading radiochromic films and related method of operation.

Technical Field The present invention relates to the general field of ionizing radiation metrology.

It relates more particularly to the reading of radiochromic films. A radiochromic film is a 2-dimensional dosimeter considered as an ideal detector with properties allowing it to be used for any type of application involving ionizing radiation. In practice, a radiochromic film is composed of monomers which, under the action of ionizing radiation, will polymerize and create a darkening of the film, directly correlated to the amount of radiation to which it has been subjected.

Prior technique

Ionizing radiation is now widely used in many applications, such as in the medical field for example.

Monitoring the radiation dose delivered to patients during these medical procedures is of great importance. More particularly, the delivery of the dose during an external radiotherapy treatment must be controlled beforehand in a personalized manner. This verification is particularly critical in radiotherapy techniques using small radiation fields and high dose levels.

To do this, a measurement using a matrix of detectors in a phantom simulating the patient is systematically carried out and compared with the provisional calculation carried out upstream as part of a treatment plan proposed by the medical staff. Another approach consists in directly obtaining the two-dimensional map of the dose distribution from the irradiation of a radiochromic film.

Currently, the most commonly used radiochromic films are of the GAFchromic™ brand, and in particular the EBT3 and EBT-XD models. The use of this dosimeter is nevertheless under-exploited today. Quantification of absorption change with dose must be reliable and accurate, which is problematic when using commercial scanners as 2D image densitometers.

Radiochromic films are currently read with commercial photographic scanners such as those marketed under the names Epson® 10000XL, 11000XL, 12000XL, V700, V750Pro or V800.

A large number of scientific publications have highlighted the difficulties of use, reading artifacts and the inadequacy of the flatbed photographic scanner to obtain reliable dosimetry and ensure the metrological traceability required for all other dosimetric detectors: [1] , [2], [3], [4], [5], [6]

These scanners were originally developed for photographic applications and are not well suited for scientific applications.

The associated software is a kind of black box that often hides useful information about the digital processing applied to the images for better visual rendering.

Moreover, these optical systems, each comprising a photographic scanner and its associated software, are not measuring instruments and are therefore not suitable for scientific applications.

The existing optical systems for reading radiochromic films can essentially be classified into three categories:

- X-ray microdensitometer systems with CCD camera (Anglo-Saxon acronym for "Charge-Coupled Device"), as described in patent US5623139;

- flatbed scanners;

- laser densitometry systems (LDS acronym for “Laser Densitometry System”).

Patent US9128053B2 and the related publication [6] describe such an LDS system which consists of the use of a laser coupled to a photodiode to perform point-by-point scanning of the radiochromic film.

The inventors have made an inventory of the main characteristics displayed and their drawbacks, summarized in the following table 1. [Table 1]:

If you are specifically interested in laser densitometry systems, the high time needed to acquire an image (2 min for a 5x5cm ² film and a resolution of 72dpi) makes it difficult to transpose it to a hospital environment for routine clinical use. The use of a laser brings up issues of speckle, shape and overlapping of the spot.

Scanning increases the acquisition area but causes overlapping and mosaic problems. Scanning during digitization can also generate geometric distortions. Lasers are more expensive and light intensities are more difficult to control. It is difficult to ensure uniform illumination of the film when scanning the area.

There is therefore a need to improve optical systems for reading radiochromic films, in particular in order to overcome the drawbacks of X-ray microdensitometers with CCD camera, flat optical scanners and existing laser densitometers.

The general object of the invention is then to meet this need at least in part.

Disclosure of Invention

To do this, the invention firstly relates to an optical system for reading a radiochromic film, comprising:

- a light source comprising at least one light-emitting diode (LED) arranged to illuminate the radiochromic film, the emission wavelength of each LED diode being adapted to correspond to that of the absorption peak(s) ( s) principal(ux) of sensitivity of the radiochromic film,

- a bi-telecentric lens arranged to receive the optical beam emitted by the diode(s) (LED) and having passed through the radiochromic film,

- an imager optically coupled to the bi-telecentric lens.

By “imager” is meant here and in the context of the invention, the usual technological meaning, namely any device with at least one photographic sensor which converts electromagnetic radiation from at least one LED into an analog electrical signal. This signal is then amplified, then digitized by an analog-digital converter and finally processed to obtain a digital image. An imager according to the invention can be at least one photographic sensor implementing any technology. It may be a CCD sensor (acronym for “Charge-Coupled Device”) or a CMOS sensor (acronym for “Complementary Metal-Oxide-Semiconductor”).

Advantageously, the imager is an sCMOS (acronym for “scientific Complementary Metal-Oxide-Semiconductor”) imager.

Preferably, the light source comprises at least one matrix of LED diodes. The light source can have high homogeneity and high intensity, typically up to 35500 mcd per LED array. In addition, one or more, in particular three, diffusing glass plates is (are) arranged above the LEDs to eliminate the hot spots of the lights emitted by the individual LEDs. The diffusing glass plates homogenize the lights into a single emission profile.

Preferably again, each LED diode has an emission length in the red, providing a quasi-monochromatic light which guarantees with the bi-telecentric lens the absence of geometric and chromatic distortions. In particular, the surface treatment of the bi-telecentric lens avoids chromatic aberrations, i.e. focusing at different distances depending on the wavelength.

The system comprising at least one suitable current controller for each row of the LED array and a sufficient voltage supply to make the currents constant for the entire LED array. Thus the output light is constant for all the LED matrices of the light source. Advantageously, the LEDs can be mounted in parallel with, for each row, a current controller which makes it possible to absorb voltage variations such that at the output, the intensity is constant for each row, and therefore for each LED of this row. . In addition, putting one controller per row makes it possible to have a more uniform intensity of all the LEDs.

According to an advantageous embodiment, the imager is a monochrome camera comprising sCMOS sensors.

According to this mode, and an advantageous embodiment variant, a microlens is arranged in front of each pixel of an sCMOS sensor of the camera. With microlenses, the effects of optical crosstalk are minimized.

According to an advantageous configuration, the optical system comprises: - a support transparent to the wavelength of the diode(s) (LED), the support being adapted to support the radiochromic film and arranged to allow the optical beam emitted by the diode(s) (LED) to pass ,

- a compression plate with an optical index close to that of the radiochromic film, the plate being mounted to be applied against the support so as to flatten the radiochromic film.

The compression plate, in particular made of anti-Newton glass, makes it possible to guarantee the flatness of the radiochromic film during the image. By "anti-Newton glass" is meant here and in the context of the invention a frosted glass whose surface is not perfectly smooth so as to leave a presence of air between the compression plate and the radiochromic film, preventing the formation of interference fringes called Newton's rings.

According to another advantageous embodiment, the system comprises an enclosure impervious to external light in which are housed at least the light source, the bi-telecentric lens, and if necessary the support with the compression plate.

Preferably, the walls of the enclosure are made of thermally insulating material.

The system includes a thermal control device for controlling the temperature of the light source.

The thermal regulation device is advantageously a heat exchanger connected to a circulation thermostat. Such a device makes it possible to obtain chromatic and thermal stability of the light source.

The invention also has a method of operating the optical system as described previously, according to which the piloting of the LED diode(s) is independent of that of the camera.

According to an advantageous embodiment, the method comprises the following steps:

- control of the LED diode(s) in pulsed mode,

- acquisition of images by the camera while the radiochromic film is illuminated by the LED diode(s).

The film is digitized with an exposure time of a few ms and an average can be obtained by acquiring several images at a high frame rate. In practice, the exposure time is managed by the user so as not to saturate the sCMOS sensor, in order to be able to measure optical densities. Indeed, a saturation of the sensor does not make it possible to know the quantity of incident light.

The global shutter of the sCMOS sensor needs external light control. In the context of the invention, two main modes of shuttering of the sCMOS sensor can be envisaged:

- the first consists in carrying out a line reading but ensuring that each line does not have the same quantity of light received,

- the second is that the entire surface of the sensor is exposed and read simultaneously. This second mode requires flash type lighting and protection against external light. If the flash dominates the exposure and the external light is suppressed, each pixel receives the same amount of light. The reading per line as well as the different exposure times of the lines no longer have any importance, since the external light no longer penetrates.

The inventors believe that this second mode of synchronizing the light source with the flash sCMOS sensor is preferable, with the following additional advantages:

- the fact of generating pulses of light makes it possible to dissipate less heat than in the case of continuous light emission. Cooling is therefore easier to achieve,

- the lifetime of the light source is increased because it operates only a few ms per frame.

According to an advantageous variant embodiment, the acquisition step comprises the acquisition of correction images called offset, flat and dark (respectively “offset”, “fiat” and “dark”).

According to another advantageous embodiment variant, the method comprises a pre-processing step by acquiring correction images in order to obtain reference images called master images of the film before irradiation and after irradiation.

Preferably, the method comprises a processing of the master images of the film comprising the following steps:

- conversion into optical density of the pixel values of the master images of the film before irradiation and after irradiation, - correction of the optical density of the master images of the film before irradiation and after irradiation by means of calibrated neutral filters so as to obtain the images of the film calibrated in optical density, before irradiation and after irradiation,

- sub-pixel registration of the master images of the film before irradiation and after irradiation,

- subtraction of the two images of the film, calibrated in optical density before and after irradiation, taking into account the image of the film before irradiation as background noise,

- application of a calibration curve to images established from films, irradiated at known doses traceable to established primary references, so as to obtain the dosimetry of the film.

Any organization with appropriate irradiation means and calibrated dosimetric measurement equipment can produce such a calibration curve for radiochromic films.

It is the realization of a calibration curve of the optical density according to the dose absorbed in a phantom, generally in water or equivalent under conditions in which the dose can be traced metrologically, and its application to the subsequent measurements carried out with films from the same manufacturing batch, which makes it possible to obtain dosimetric information from the films.

Thus, the invention essentially consists of an optical system comprising a uniform LED source, an imager, advantageously an sCMOS imager, preferably monochrome, coupled to a bi-telecentric optical lens.

The use of a bi-telecentric lens makes the imager insensitive to film polarization.

The method of operation of the system allows the determination of the dose after irradiation of the film by measuring the two-dimensional optical density on the film.

The use of calibrated neutral filters makes it possible to restore metrological traceability in optical density compared to flatbed scanners according to the state of the art. In other words, the invention allows reliable measurements of the optical density of radiochromic films.

By construction, flatbed optical scanners according to the state of the art do not allow reliable measurements of optical densities for various reasons, including the use of incident white light, the inhomogeneity of the light source, the saturation of the sensor and internal recalibration.

By using dose measurements traceable to primary references established by an organization, as well as calibrated neutral filters, the system according to the invention is a fully metrologically traceable instrument whose performance can be quantified in terms of uncertainties. Its performance greatly exceeds that of state-of-the-art optical flatbed scanners.

The advantages of the invention are numerous, among which we can cite:

- a dedicated scientific instrument, sized for reading radiochromic films, in order to ultimately have reliable dosimetric information, while obtaining double metrological traceability of the optical density and the radiation dose,

- the implementation of the optical components of the system, the characteristics of which can be controlled precisely and independently, makes it possible to avoid reading artifacts known with existing flatbed scanners, with rigorously qualified metrological performance,

- the system is large enough to reproduce the full size of a 6x6 cm ² film, or even larger dimensions if the support is movable on two axes, the sCMOS imager having a resolution carefully adapted to the objective to provide a size ideal pixel range on film to capture fine detail in the exposed pattern,

- a very stable instrument thanks to the addition of current controllers and reproducible by the periodic calibration of the system in optical density,

- independent control of each optical component of the system allows greater flexibility for acquisition (image sequences, exposure time control) and image processing,

- a real quantification of the uncertainties on the result is achievable, which is currently lacking with flatbed scanner systems according to the state of the art.

The invention can be considered for all applications and industrial fields implementing 2D dose measurements in medical applications of ionizing radiation, as well as for the estimation of the dose to patients for the characterization of the radiation field.

The invention can even be implemented to reconstruct a dose in 3D, by stacking several radiochromic films alternately with slices so as to reconstitute a phantom, the reconstruction taking place by interpolation between the slices.

For example, a linear accelerator, X-ray generators or isotopic sources can generate these radiation fields not exclusively.

A particularly interesting application of the invention is for medical physicists in radiotherapy or radiology departments and for any producer of apparatus and instruments in the medical field.

Other advantages and characteristics will emerge better on reading the detailed description, given by way of illustration and not limitation, with reference to the following figures.

Brief description of the drawings [Fig 1] Figure 1 is a perspective and partially exploded view of an optical system for reading radiochromic films according to the invention.

[Fig 2] Figure 2 is a longitudinal sectional view of part of the optical system according to Figure 1.

[Fig 3] Figure 3 is a block diagram of the protocol for preprocessing images obtained using an optical system according to the invention.

[Fig 4] Figure 4 is a block diagram of the sequence of the image processing process obtained using an optical system according to the invention.

[Fig 5] Figure 5 is an illustration of a radiochromic film irradiated by a beam of photons from an optical system according to the invention. detailed description

Throughout the application, the terms "top", "bottom", "top", "bottom" are to be considered with a system according to the invention in a vertical configuration, i.e. the optical axis of the components is vertical and the radiochromic film to be read horizontally on its support. Figure 1 shows a system 1 for optical reading of radiochromic films according to the invention. System 1 extends along a longitudinal axis X.

The system 1 comprises an enclosure 10 impermeable to external light mounted on a table 11. Thus, the walls of the enclosure 10 are made of a material making it possible to suppress any external light and any parasitic reflection. The walls of enclosure 10 are coated at least one thermally insulating material, in order to thermally insulate the interior of the enclosure from the external environment. For example, the walls 10 can be lined on the inside of the enclosure with black flocked paper and on the outside with aluminized polystyrene.

System 1 comprises a light source 2 comprising at least one light-emitting diode (LED) arranged to illuminate a radiochromic film. It can be a matrix of LEDs. This light source 2 makes it possible to obtain a uniform light, for example in the red.

The wavelength of each LED is chosen to coincide with the maximum sensitivity absorption peak of the radiochromic film to be read.

Such a source 2 is high efficiency due to the large number of LEDs it is possible to have in a matrix, which maximizes the signal-to-noise ratio and reduces the exposure time.

Preferably, current controllers for each LED matrix are provided with a sufficiently high voltage supply to make the currents constant and therefore the light output constant for all the LED matrices.

The matrix of LEDs 2 is arranged inside a support 20 transparent to the wavelength of the diode(s) (LED), which is adapted to support the radiochromic film and arranged to allow the optical beam emitted to pass by diode(s) (LED) 2.

A compression plate 21, with an optical index close to that of the radiochromic film, is mounted to be applied against the support so as to flatten the radiochromic film. This plate 21 can be made of anti-Newton glass. The compression provided by the plate 21 to flatten the radiochromic film makes it possible to avoid the variation in intensity of the pixels with the curvature of the film (Callier effect).

As shown in Figure 2, three diffusing glass plates 22 are arranged above the LEDs while being held in the support 20. These plates 22 make it possible to eliminate hot spots. A bi-telecentric lens 3 in the form of an objective is arranged directly above the support 2 to receive the optical beam emitted by the diode(s) (LED) 2 and having passed through the radiochromic film,

Such a lens 3 makes it possible to select the light rays almost parallel to the optical axis in the object space (radiochromic film) and the image space (sCMOS sensor). Thereby, one can have a maximum angle of incidence of the light to the sensor, typically of the order of 1.1°.

The coating of the lens 3 is preferably treated to avoid chromatic aberrations, since this phenomenon exists even for monochromatic light and a narrow emission peak.

Also, the edges of the lens 3 are blackened and a device is provided to avoid parasitic reflections.

An sCMOS imager 4 with a CMOS sensor is aligned with the lens 3 to receive the optical beam having passed through the latter. The pixel size of a sensor can be equal to 6.5um.

The imager 4 is preferably cooled and regulated by the Peltier effect.

Each sensor pixel is preferably covered by a microlens. This increases the quantum efficiency of the collection and decreases the effect of crosstalk with neighboring pixels.

Such an sCMOS 4 imager is monochrome to avoid signal loss and artifacts in the so-called Bayer filter array and at high resolution, typically 6.5 µm per side (2048*2048 pixels).

The sensors implemented have a high dynamic range leading to a very large dynamic range of image absorption measurement. In addition, an sCMOS camera has low thermal and readout noise.

The sCMOS imager 4 and the bi-telecentric lens 3 are fixed to a flange 11 itself fixed to a bracket 12.

As shown in figure 1, the LED matrix light source 2, the bi-telecentric lens 3, and the support 20 with the compression plate 21 are housed inside the enclosure 10.

In order to maintain a constant temperature for the source 2 of LEDs, a thermal regulation device 5 is installed.

This device is preferably constituted by a heat exchanger connected to a circulation thermostat. With such a device 5, chromatic and thermal stability is obtained at inside source 2 of the LEDs. The temperature is adjusted to synchronize the LED emission peak with the maximum absorption peak of the radiochromic films to be read.

The method of operation of such a system is as follows.

The light source 2 and the sCMOS imager 4 are independently controlled.

The bi-telecentricity of the lens 3 makes the imager 4 insensitive to the polarized light coming from the radiochromic film: the difference between the S- and P- polarizations depends on the angle of incidence of the polarized light.

System 1 operates by driving the LED diode(s) in pulsed mode and synchronizing the image acquisition with the triggering of the diode(s).

Pulsed mode operation allows the radiochromic film to be illuminated only during image acquisition, minimizing exposure of the film to light and enabling use of the global shutter mode of operation of an sCMOS sensor.

Another advantage of pulsed mode is to limit unnecessary heat generation.

Image acquisition and preprocessing protocol:

The system 1 which has just been described allows an improvement of the reading protocol used until now thanks to the independent control of the sCMOS imager 4 and of the light source 2 by being able to change the exposure time.

With independent driving, multiple images can be taken at high speed, thereby reducing the statistical uncertainty in the amount of radiation dose measured by the radiochromic film.

The acquisition of three types of complementary corrective images, called shifted, dark, flat. This acquisition of images can be implemented by software produced in Python language using the libraries and the computer drivers made available by the manufacturer of the camera 4, such as for example the manufacturer PCO Imaging.

A preprocessing with these corrective images is carried out, in order to obtain correct quantitative information in reference images (masters).

The preprocessing sequence is explained in Figure 3. This preprocessing can be implemented by software produced in Python language using different libraries freely accessible mathematical or image processing, such as: https://matplotlib.org/, https://numpy.org/, https://www.scipy.org/, https://opencv.org/ .

This preprocessing can help eliminate problems with lot-to-lot variation calibration of radiochromic films. Also, it compensates for the curvature of the light source, with all lights having a limited aperture area of illumination.

Image processing protocol:

Pixel values from master images are converted to optical density after pre-processing and then corrected from calibrated neutral filters, resulting in traceable optical density radiochromic film images.

A double reading of each radiochromic film before and after irradiation is systematically carried out, to take into account the intrinsic variations of the dosimeter, i.e. the possible inhomogeneity of the film.

A sub-pixel registration based on the recognition of the four corners of the film and the calculation of a geometric transformation are carried out, which makes it possible to register the non-irradiated film and the irradiated film to subtract these recordings in a second time, the non-irradiated film irradiated being used as background noise image. The registration can be a rigid registration which only requires rotation/translation operations. Here, the film has not been dimensionally transformed so a rigid registration is sufficient. The application of a calibration curve to the images established from films, irradiated at known doses traceable to established primary references, and the optical densities measured with the optical system 1 which has just been described, makes it possible to recover dosimetric information.

The complete processing sequence is shown in Figure 4. An example of radiochromic film imaged using the described system and method of operation is shown in Figure 5.

The invention is not limited to the examples which have just been described; it is in particular possible to combine together characteristics of the examples illustrated within variants not illustrated. Other variants and improvements can be envisaged without departing from the scope of the invention.

For example, with a fixed support, it is possible, thanks to the optical system which has just been described, to read radiochromic films of dimensions up to 6×6 cm ² . By adapting a movable support (motorized) on two axes, it is possible to image films of larger dimensions per portion. The entire image of the film can then be reconstructed by concatenating the partial images acquired.

In the detailed example, the matrix of LEDs is preferentially chosen to emit exclusively in the red. In the context of the invention, it is quite possible to envisage having a mixed light source with LEDs emitting in the red and others in the green corresponding to the main absorption emission peaks and secondary of the gafehromic type radiochormic films.

List of cited references:

[1]: Niroomand-Rad A, Chiu-Tsao ST, Grams MP, et al. “Full report of AAPM Task Group 235 radiochromic film dosimetry: An update to TG-55 Med Phys. 2020;10.1002/mp.14497. doi: 10.1002/mp.14497.

[2]: Poppinga D, Schoenfeld AA, Doemer KJ, Blanck O, Harder D, Poppe B. “A new correction method serving to eliminate the parabola effect of flatbed scanners used in radiochromic film dosimetry ” Med Phys. 2014;41(2):021707. doi: 10.1118/1.4861098.

[3]: Schoenfeld AA, Poppinga D, Harder D, Doerner KJ, Poppe B. “77ie artefacts of radiochromic film dosimetry with flatbed scanners and their causation by light scattering from radiation-induced polymers.” Phys Med Biol. 2014;59(13):3575-3597. doi: 10.1088/0031-9155/59/13/3575

[4]: Lewis D, Chan MF. “ Correcting lateral response artifacts from flatbed scanners for radiochromic film dosimetry. “Med Phys. 2015;42(l):416-429. doi: 10.1118/1.4903758

[5]: Simcoe RJ. “Evaluating Commercial Scanners for Astronomical Image Digitization in Preserving Astronomy's Photography Legacy: Current State and the Future of North American Astronomical Plates. ” ASP Conference Series, Vol. 410. Edited by Wayne Osborn and Lee Robbins. San Francisco: Astronomical Society of the Pacific, 2009., p.111. [6]: Rosen BS, Soares CG, Hammer CG, Kunugi KA, DeWerd LA. “A prototype, glassless densitometer traceable to primary optical standards for quantitative radiochromic film dosimetry ” Med Phys. 2015;42(7):4055-4068. doi:10.1118/1.4922134

Claims

1. Optical system (1) for reading a radiochromic film, comprising:

- a light source (2) comprising at least one light-emitting diode (LED) arranged to illuminate the radiochromic film, the emission wavelength of each LED diode being adapted to correspond to that of the peak(s) d main sensitivity absorption(s) of the radiochromic film,

- a bi-telecentric lens (3) arranged to receive the optical beam emitted by the diode(s) (LED) and having passed through the radiochromic film,

- an imager (4) optically coupled to the bi-telecentric lens.

2. Optical system (1) according to claim 1, the light source comprising at least one matrix of LED diodes.

3. Optical system (1) according to claim 1 or 2, each LED diode having an emission length in the red.

4. An optical system (1) according to claim 2, comprising, at least one current controller suitable for each row of the LED array and a voltage supply sufficient to make the supply currents constant for any LED array. .

5. Optical system (1) according to one of the preceding claims, the imager being a so-called sCMOS monochrome camera.

6. Optical system (1) according to claim 5, comprising a microlens arranged in front of each pixel of a CMOS sensor of the camera.

7. Optical system (1) according to one of the preceding claims, comprising:

- a support (20) transparent to the wavelength of the diode(s) (LED), the support being adapted to support the radiochromic film and arranged to allow the optical beam emitted by the diode(s) to pass (LED),

- a compression plate (21), with an optical index close to that of the radiochromic film, the compression plate being mounted to be applied against the support so as to flatten the radiochromic film.

8. Optical system (1) according to one of the preceding claims, comprising an enclosure impervious to external light in which are housed at least the light source, the bi-telecentric lens, and if necessary the support with the compression plate .

9. Optical system (1) according to claim 8, the walls of the enclosure being made of thermally insulating material.

10. Optical system (1) according to claim 8 or 9, comprising a thermal regulation device (5) for regulating the temperature of the light source.

11. Optical system (1) according to claim 10, the thermal regulation device being a heat exchanger connected to a circulation thermostat.

12. Method of operating the optical system according to one of the preceding claims, according to which the control of the LED diode(s) is independent of that of the imager.

13. Operating method according to claim 12, comprising the following steps:

- control of the LED diode(s) in pulsed mode,

- acquisition of images by the imager during the illumination of the radiochromic film by the LED diode(s).

14. Method of operation according to claim 13, the acquisition step comprises the acquisition of correction images called shifted, flat and dark.

15. Operating method according to claim 14, comprising a pre-processing step in order to obtain reference images called master images of the film before irradiation and after irradiation.

16. Operating method according to claim 15, comprising a processing of the master images comprising the following steps:

- conversion into optical density of the pixel values of the master images of the film before irradiation and after irradiation,

- correction of the optical density of master images of the film before irradiation and after irradiation by means of calibrated neutral filters, so as to obtain the images of the film calibrated in optical density, before irradiation and after irradiation,

- sub-pixel registration of the master images of the film before irradiation and after irradiation, - subtraction of the two images of the film calibrated in optical density before and after irradiation taking into account G acquisition before irradiation as background noise,