Reconstructing dosimetric images using a deconvoluted physical model

Restoration of images acquired using planar detectors and a TL reader with CCD camera readout
694 Downloads
Updated 1 Jan 2012

View License

The development of planar (2D) dosimetry system using Laboratory TL reader was a basic goal of the work conducted under MAESTRO programme (http://www.maestro-research.org/) at the Institute of Nuclear Physics (IFJ PAN) in Krakow, Poland.
There are many grading parameters of the system among which image resolution is of unquestionable importance. Images of good resolution can be an useful tool for dosimetry of small radiation sources or beams with steep dose gradients. Because of that any image manipulation method basing on the hardware properties of the reader, leading to increase of the resolution is demanded option in the readers software.
-----------
Implementation:
This code reconstructs dosimetric images acquired using planar detectors (thermoluminescence foils of LiF:Mg,Cu,P powder and ethylene-tetrafluoroethylene (ETFE) copolymer) and a TL reader with CCD camera readout.

The algorithm creates a point-spread function (psf) based on certain physical properties of the reader: path of the light transmitted, size of the grain, distance from the CCD camera, etc. and deconvolves the original image making use of the Matlab Image Processing Toolbox. The deconvolved image is supposed to
improve the dosimetric quality of the readouts.
-----------
Input parameters:
Dia: diameter of the foil grain (in um.)
dist: distance between the lens of the CCD camera and the TLD detector (in mm.)
imgloc: input image file produced by the TL reader
alg: selected deconvolution algorithm (Lucy-Richarson='lucy';Blind='blind'; Regularized filter='reg'; Wiener filter='wnr')
optSelected: operation mode selected (=1 for standard 5cmx5cm detector size; =2 sets up 1cmx1cm operation)
nluc: number of iterations (if required by the deconvolutution algorithm)

-----------
Example usage:
The image file 'D-9.TXT' captures the dose distribution on thermoluminescence (TL) detector after irradiation by microbeam at European Synchrotron Radiation Facility. The dimension of TL detector is 1 cm square and was produced from MCP-N powder (grain-size 75-90 micrometer) and high temperature polymer. The irradiation was realised using synchrotron X-rays microbeams of 75 mm FWHM peak and 400 mm centere-to centere (c-t-c) spacing between peaks. The readout was realised by TL Reader equipped with CCD camera and Tevidon 1,4/25 mm lens. The optical setup was set in macro mode. Distance from TLD plate to the lens was 10cm.

For 15 iterations using the Lucy-Richardson deconvolution algorithm

>> [imagereader,deconvolutedNormalized] = doseImageDeconv(90,100,'D-9.TXT','lucy',2,15);

>> figure;
>> subplot(1,2,1);
>> imagesc(imagereader); title('input image');
>> subplot(1,2,2);
>> imagesc(deconvolutedNormalized); title('output image');
-----------
This work was presented at the Marie Curie Conference, satellite event of the Euroscience Open Forum ESOF 2010:

Madrigal P. (2010) Image processing in nuclear physics: Planar (2D) thermoluminescence detectors image reconstruction using two-dimensional deconvolution. Marie Curie Conference, Euroscience Open Forum ESOF 2010, 1-2 July 2010, Torino, Abstracts Book, P-219.

References on thermoluminescence dosimetry:

Marczewska B, Bilski P, Czopyk L, Olko P, Waligórski MP, Zapotoczny S. (2006) Two-dimensional thermoluminescence dosimetry using planar detectors and a TL reader with CCD camera readout. Radiat Prot Dosimetry 120(1-4):129-32.

Olko P, Marczewska B, Czopyk L, Czermak MA, Klosowski M, Waligórski MP. (2006) New 2-D dosimetric technique for radiotherapy based on planar thermoluminescent detectors. Radiat Prot Dosimetry 118(2):213-8.

Czopyk L, Klosowski M, Olko P, Swakon J, Waligorski MP, Kajdrowicz T, Cuttone G, Cirrone GA, Di Rosa F. (2007) Two-dimensional dosimetry of radiotherapeutical proton beams using thermoluminescence foils. Radiat Prot Dosimetry 126(1-4):185-9.
-----------
The author gratefully acknowledges Pawel Olko, Lukasz Czopyk, Mariusz Klosowski & Jan Gajewski

Cite As

Pedro Madrigal (2024). Reconstructing dosimetric images using a deconvoluted physical model (https://www.mathworks.com/matlabcentral/fileexchange/34407-reconstructing-dosimetric-images-using-a-deconvoluted-physical-model), MATLAB Central File Exchange. Retrieved .

MATLAB Release Compatibility
Created with R2007b
Compatible with any release
Platform Compatibility
Windows macOS Linux

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!
Version Published Release Notes
1.0.0.0