A VIRTUAL X-RAY IMAGING AND PROCESSING SYSTEM FOR PHOTON ABSORPTION PROCESSES

F. MALANO; C. QUINTANA; G. TIRAO; M. VALENTE

Tandil

Workshop; 2da Escuela y Workshop de Ciencias de las Imágenes – ECImag; 2009

Nowadays, X-ray imaging is worldwide one of the most used non-invasive techniques for medical applications. Actually, conventional and digital radiography, mammography and computed tomography, three of the most commonly employed techniques are based on similar physical features, which involve radiation-matter interaction processes. Medical, physical, biochemical and even metabolic quantities are traduced into images, which may be suitably processed in order to optimize relevant information extraction. Therefore, image processing techniques have become great interests for medical imaging during the last years. However, due to intrinsic complexity, difficulties regarding the access to modern medical imaging facilities as well as high costs, direct measurements are sometimes hard to perform. On the other hand, virtual techniques, like Monte Carlo simulations have proved to be a useful tool to this aim. Actually, Monte Carlo simulations have a wide range of applications in radiation transport. This work presents an original development consisting on a complete stochastic calculation routine (IMAGING program based on the main code PENELOPE) devoted to X-ray imaging, which is suitably integrated with a specially adapted system (XRIA program MatLab® supported) for simulation output data acquisition and processing in order to assess image quantification. A user-friendly graphical interface (developed with GUI MatLab® environment) allows to load, process and visualize -automatically or manually- the simulated virtual images. The developed system has been applied to the investigation of the energy spectrum effect on the X-ray image quality for biological materials (low effective atomic number). In this sense, different X-ray tube anode materials (Molibdenum, Tungsten and Cobalt) as well as different accelerating voltages, ranging from 20 to 60 kV according to conventional X-ray tubes, have been considered. Image (absorption) contrast defined as intensity differences has been proposed as relevant parameter for image quality characterization. The obtained results show a strong image quality dependence upon the incident beam energy spectrum. Actually, it was possible toestablish the optimal incident beam according to sample characteristics, which represents a really significant feature in practical cases. Furthermore, the developed integrated Monte Carlo routine along with the image processing system have proved to constitute a useful tool, even showing significant advantages with respect to traditional methods, for X-ray imaging investigations.