IFEG   20353
INSTITUTO DE FISICA ENRIQUE GAVIOLA
Unidad Ejecutora - UE
congresos y reuniones científicas
Título:
X-Ray Fluorescence Scanning of Human Teeth and Dedicated Image Processing
Autor/es:
R. FIGUEROA; A. DIAZ; M. VALENTE
Lugar:
Porto Alegre
Reunión:
Congreso; 18th International Congress of Medical Physics; 2011
Resumen:
Nowadays, X-ray fluorescence constitutes a useful and accurate technique, worldwide established for constituent elementary distribution assessment. Actually, concentration distributions of arbitrary user-selected elements can be achieved along sample surface with the aim of identifying and simultaneously quantifying every constituent element. Visualization of elemental distributions of biological sample is gaining importance in many disciplines of biological, forensic, and medical research. Images are prompt achieved (within few minutes) scanning with a specially designed robotic arm which performs step by step or continuous positioning. The scanning areas considered in this study are for samples up to 100 cm2 along with variable spatial resolution capable of reaching 0.2 mm2 per pixel. XRF spectra are detected at each position (x,y) in times of up to 1ms. Dedicated image processing programs ? MatLab® supported - have been developed with the aim of performing suitably automatic XRF image analysis. Furthermore, developed programs are integrated by means of dedicated graphic interface for user interaction. The integrated tool allows to process and analyze acquired XRF images by means of different techniques including noise filtration, background subtraction, selective spectrum peak individualization, single/multiple element mapping and even image fusion. After preliminary tests, the whole XRF scanning system has been used for different applications. This work resumes the main characteristics of the new portable XRF scanning system. In addition, system performance is investigated for applications to human tooth obtaining. Dental sample elemental spatial distributions are reported and discussion is presented regarding the use of this information for further correlation with the corresponding pathologies associated.