RU2529454C1 - Method for experimental determination of point blurring function in converter for detecting proton radiation - Google Patents

Abstract

FIELD: physics.

SUBSTANCE: method includes outputting and deviating from a proton beam a portion of protons by passing said protons through a curved channelling crystal and directing towards a converter of a detecting system. A region of localisation of transmitted protons and a region formed by scattered protons and secondary particles are selected on the obtained image. Further, the size of the image of the reconstructed point blurring function is calculated in the form a square and then, by varying the pixel value in the region of localisation of transmitted protons and the square region, a set of pixel values is selected in the square region, which is the reconstructed point blurring function.

EFFECT: high reliability of information when determining a point blurring function.

3 cl, 5 dwg

Description

The invention relates to the field of proton radiography, in particular to methods for recording optical images formed using proton radiation, and can be used, for example, in digital recording systems to determine the internal structure of objects or to study fast processes.

The task in this field of technology is to obtain reliable information about the properties of the recording system for optimizing proton complexes and verifying calculation methods when modeling the processes of radiation passing through matter.

When forming an image of the studied region (object) using a proton beam, a converter (scintillation transducer, hereinafter - scintillator) of the proton beam energy in the optical range and its transfer system to input image intensifier tube. In a scintillator (plastic, LSO, etc.), a high-energy proton as a result of strong and electromagnetic interactions leaves part of its energy, and the released energy is converted into information carriers (quantum of light, free charge carriers, etc.) that can be processed and recorded optical and electronic devices.

Protons from the source for scanning through the study area are fed in clots at strictly defined time intervals, which allows several frames to be taken. When registering images in proton radiographic experiments, an error appears due to the imperfect nature of the recording system. During the interaction of protons with the scintillator matter, the atoms of the substance ionize along the proton trajectory; in addition, 5-electrons with significant ranges can be formed, as well as secondary particles - the result of inelastic nuclear interaction of high-energy protons with the scintillator material nuclei. As a result, the energy released when a proton enters the scintillator is localized not only near the proton trajectory, but can also be located at a considerable distance from its trajectory. This circumstance leads to the fact that a light burst from a single proton entering the scintillator is recorded not as a point, but as a certain distribution. When averaging these outbreaks over a large number of incident protons, a smoothed, noise-free function is obtained, called the point blur function (PSF), which should be taken into account when processing data to restore the original image without error.

The traditional method of experimental production of PSF in radiography using neutral particles (radiography, neutron diffraction) is to measure the function of blurring the face (Germany) [V. Masschaele et al. / Nuclear Instruments and Methods in Physics Research A 542 (2005) 361-366, UCRL-JC-130427, Preprint, Roll Bar X-Ray Spot Size Measurement Technique, R.A. Richardson and T.L. Houck / or / OPTICAL DIAGNOSTICS ON ETA II FOR X-RAY SPOT SIZE, R.A. Richardson, Proceedings of the 1999 Particle Accelerator Conference, New York, 1999]. Having Germany, you can easily calculate the line blur function (FDR), and then the FRT.

However, in the case of proton radiography, using charged particles as translucent particles, as well as magnetic optics to compensate for the scattering of protons in the material of the object, it is impossible to accurately distinguish blurring due to chromatic aberrations in the final image.

It is also difficult to obtain a PSF, due to the following factors.

First, in real scintillators, the light output is a function of the energy density. For example, for organic scintillators, this dependence is described by the Birks semiempirical model [Bartels J., Haidt D., Zichichi A., The European Physical Journal C, v.15, 2000], the coefficient in which must be obtained experimentally for each scintillator, and far not for all scintillators, this coefficient is measured and published.

Secondly, when modeling processes, the question arises of the accuracy of modeling the interaction of particles with the medium.

Thirdly, there are factors associated with optical effects, among which there is optical blurring by lenses, associated with the thickness of the scintillator and the fact that light is emitted not from the plane, but from the entire volume of the scintillator. Also, for the correct simulation, it is necessary to have experimental data on the quality of scintillator surface treatment and on the volume scattering of light in the scintillator material.

Thus, to take into account the optical processes of light transfer, it is necessary to carry out a series of special experiments to study these issues for the experimental reconstruction of PSF. Without this blur function, the points obtained by calculation have only an approximate form. The inventive method does not have a prototype, because from the prior art there are no similar methods for experimentally determining the PSF in a converter for detecting proton radiation.

The technical result of the proposed method is to increase the reliability of information in determining the PSF.

The specified technical result is achieved due to the fact that the FTF in the converter for detecting proton radiation is restored experimentally, for which part of the protons is removed and deflected from the proton beam by passing them through a curved channeling crystal, one of whose ends is arranged towards the proton flux, directing the extracted protons to the converter of the recording system, after which the region of localization of the extracted protons, depending on the experimentally obtained image, is determined south of the transverse size of the crystal and its location relative to the converter, and according to the brightness of the pixels, the region formed by scattered photons and secondary particles resulting from the interaction of protons with the converter substance is distinguished, then the image size of the reconstructed PSF in the form of a square is calculated from the image of the last region, and then by varying the pixel values in the region of localization of the transmitted protons and the square region, a set of pixel values in the square region is selected at which p convolution result image localization and square region most closely matches the experimentally obtained image. This set is a restored PSF.

The crystal may be made of silicon in the form of an elongated curved plate.

The selection of a set of pixel values in a square region for convenience may be limited by the condition of central symmetry.

The inventive method is based on the use of such a phenomenon as the channeling of high-energy protons by curved crystals [A.G. Afonin et al. / Nucl. Instr. and Meth. in Phys. Res. In 234 (2005) 14-22]. Using this phenomenon and choosing the appropriate crystal shape, one can obtain protons localized in one region.

The use of a curved channeling crystal, one of whose ends is opposite the proton flux, directing the extracted protons to the converter of the recording system, allows localizing the interaction of protons with the scintillator in a small region with sharp boundaries.

To calculate the PSF, it is taken into account that protons deflected by a crystal fall only in a limited region of the scintillator formed by protons trapped and deflected by the crystal, as well as decanalized protons. The dimensions of this region are estimated based on the cross section of a bent crystal, and also taking into account the scattering of protons in the air gap between the crystal and the scintillator.

Figure 1 shows a diagram of an experiment to restore PSF; figure 2 shows the image obtained on an LSO scintillation plate of the image of protons deflected by a crystal oriented end to end towards the flow; figure 3 - part of the image passed through the crystal of protons, used in mathematical processing; figure 4 is a profile of the restored PSF; figure 5 is a comparison with the calculated profile.

The experimental design for the proposed method is as follows. The proton beam exiting the ion guide 1 at the image recording location where the lens and the CCD camera are located is cut off by a lead collimator 3 and passes through the channeling crystal 4. At the same time, a small part of them is deflected and focused using a mirror 5 on the converter in the form of a thin rectangle corresponding to the end face of the crystal. An LSO 6 scintillator was used as a converter. The experiments were carried out with various silicon crystals made in the form of rectangular plates with a bend (providing an angle of proton deflection).

The measurements were carried out as follows. Crystal 4 was oriented to the "beam" position, i.e. along the flight paths of protons, while some of the protons were captured, removed by crystal 4 from the beam and recorded using scintillator 6. Figure 2 shows the image of deflected protons recorded on scintillator 6. In this figure, a bright spot corresponding to protons captured by crystal 4 is visible , as well as a broad line due to the so-called decanalized protons, i.e. protons captured by the crystal, but removed from it due to defects in the crystal lattice. Figure 3 presents a portion of the proton image obtained on a plastic scintillator with a thickness of 10 mm, with dimensions on a logarithmic scale, after subtracting the background illumination in gray gradations. Information from this part is used for further mathematical analysis.

To calculate the PSF, we assume that the protons deflected by the crystal fall only in a limited region of the scintillator 6, which is formed by 4 protons trapped and deflected by the crystal, as well as decanalized protons. The dimensions of this region are estimated based on the cross section of the bent crystal 4, and also taking into account the scattering of protons in the air gap between the crystal and the scintillator 6. The boundary of this region is shown in Fig. 3 by a dashed line. Let us describe the technique for reconstructing PSF. Suppose we have an experimentally recorded image of a crystal with a subtracted background from secondary particles. For processing from this image, a rectangular region Ω _exp (shown in Fig. 3) is selected, within which the registered signal is significantly higher than the background signal. Next, the boundaries of the region of _cristae Ω are calculated, into which protons captured by the crystal fall (marked by a dashed line in Fig. 3). The selection area of PSF in the form of a square matrix is also selected; the dimensions of this region are selected based on the dimensions of the region Ω _exp . As an initial approximation of the PSF matrix, a δ-function is selected, i.e. the central element of the matrix is 1, the rest are 0. The algorithm for selecting PSF is as follows. At the same time, the values of pixels in the region of _Ωcrist and the values of the PSF pixels are varied, and by minimizing the function of many variables, we select an image _Imc (x, y) in the region of _Ωcrist , and such a matrix of the point scattering function so that their convolution is closest to experimentally recorded image. When varying the PSF values, the assumption of its central symmetry is used. To implement this, a program was developed that allows one to solve a class of similar optimization problems. Figure 4 shows the profile of the PSF selected in this way, figure 5 shows a comparison of it with the calculated energy release profile, where it can be seen that the calculated point blur function is very different from the experimentally reconstructed one, which clearly shows the need for experimental production of PSF for its use in processing the resulting information to increase its reliability.

Claims (3)

1. A method for determining experimentally the point blur function (PSF) in a converter for detecting proton radiation, including removing and deflecting part of the protons from the proton beam by passing them through a curved channeling crystal, one of whose ends is facing the proton flux, directing the extracted protons to the converter a recording system, after which an experimentally obtained image highlights the localization region of the extracted protons, depending on the transverse size of the crystal and its size position relative to the converter, and according to the brightness of the pixels, the region formed by scattered photons and secondary particles formed as a result of the interaction of protons with the converter substance is distinguished, then the image size of the reconstructed PSF in the form of a square is calculated from the image of the last region, and then, varying the pixel values in the localization region past protons and a square region, select a set of pixel values in a square region, at which the result of the convolution of images of the locale ation and square region most closely matches the experimentally obtained image, and it is this set of reconstructed PSF. 2. The method according to claim 1, characterized in that the crystal is made of silicon in the form of an elongated curved plate. 3. The method according to claim 1, characterized in that the selection of a set of pixel values in a square region is limited by the condition of central symmetry.

Images