Reducción de ruido en imágenes de mamografía con contraste de fases obtenida mediante haces de sincrotrón
Fecha
2019-10-03
Autores
Brizuela Cardoso, Javier David
Ruiz Gonzalez, Yusely
Perez Díaz, Marlen
Título de la revista
ISSN de la revista
Título del volumen
Editor
Resumen
La mamografía con contraste de fase (CF) usando haces de sincrotrón ha mostrado cuantitativamente mejores resultados en cuanto contraste de imagen y detección de pequeños detalles respecto a las técnicas digitales convencionales hospitalarias. Sin embargo, esta nueva técnica muestra mayores niveles de ruido aleatorio que en los equipos de mamografía digital convencionales, por lo que el objetivo del presente trabajo fue reducir ruido en dichas imágenes hasta niveles semejantes a los de los equipos hospitalarios. Se procesaron 23 imágenes de mamografías realizadas a un maniquí antropomórfico, obtenidas mediante técnicas con CF usando radiación X de sincrotrón, bajo diferentes condiciones de adquisición en cuanto a energía del haz y dosis en SYRMED, Trieste, Italia. El procesamiento de estas imágenes fue basado en filtrado clásico, bilateral y familia de la Transformada Wavelet. Se cuantificó, mediante indicadores de calidad de imagen, la relación señal a ruido (SNR), contraste (C), figura de mérito (FoM) y la visibilidad de detalle (V), así como la resolución espacial a través de la Función de Transferencia de la Modulación (MTF) en regiones de interés sobre las imágenes filtradas con las diferentes técnicas y en las originales sin filtrar. Se estudió estadísticamente el desempeño de los filtros, buscando los que propician la mejor calidad de imagen. Se apreció el aumento del indicador de calidad de imagen SNR. Con respecto a V, C y resolución espacialno fue afectado sensiblemente, lo cual demuestra la superioridad en calidad de imagen de esta novedosa técnica para la detección de pequeñas lesiones como microcalcificaciones y masas de bajo contraste.
Mammography with phase contrast (CF) using synchrotron beams has shown quantitatively better results in terms of image contrast and detection of small details compared to conventional hospital digital techniques. However, this new technique shows higher levels of random noise than in the equipment of Conventional digital mammography, so the objective of this work was to reduce noise in these images to levels similar to those of hospital equipment. 23 mammography images were performed on an anthropomorphic phantom, obtained by means of CF techniques using synchrotron X radiation, under different conditions of acquisition in terms of beam energy and dose inSYRMED, Trieste, Italy. The processing of these images was based on classic, bilateral filtering and Wavelet Transform family. The signal-to-noise ratio (SNR), contrast (C), was quantified using image quality indicators. figure of merit (FoM) and detail visibility (V), as well as spatial resolution through the Modulation Transfer Function (MTF) in regions of interest on images filtered with different techniques and in originals without filter. The performance of the filters was statistically studied, looking for those that favor the best image quality. The increase in the SNR image quality indicator was appreciated. With respect to V, C and spatial resolution, it was not significantly affected, which demonstrates the superior image quality of this novel technique for the detection of small lesions such as microcalcifications and low contrast masses.
Mammography with phase contrast (CF) using synchrotron beams has shown quantitatively better results in terms of image contrast and detection of small details compared to conventional hospital digital techniques. However, this new technique shows higher levels of random noise than in the equipment of Conventional digital mammography, so the objective of this work was to reduce noise in these images to levels similar to those of hospital equipment. 23 mammography images were performed on an anthropomorphic phantom, obtained by means of CF techniques using synchrotron X radiation, under different conditions of acquisition in terms of beam energy and dose inSYRMED, Trieste, Italy. The processing of these images was based on classic, bilateral filtering and Wavelet Transform family. The signal-to-noise ratio (SNR), contrast (C), was quantified using image quality indicators. figure of merit (FoM) and detail visibility (V), as well as spatial resolution through the Modulation Transfer Function (MTF) in regions of interest on images filtered with different techniques and in originals without filter. The performance of the filters was statistically studied, looking for those that favor the best image quality. The increase in the SNR image quality indicator was appreciated. With respect to V, C and spatial resolution, it was not significantly affected, which demonstrates the superior image quality of this novel technique for the detection of small lesions such as microcalcifications and low contrast masses.
Descripción
Palabras clave
Mamografía con Contraste - Fase, Técnicas de Filtrado, Calidad de Imagen, Phase Contrast Mammography, Digital Filter, Image Quality
Citación
Citar según la fuente original:
[1] Moeckli R, Verdun F, Fiedler S, M. et al. (2000) ―Objective comparison of image quality and dose between conventional and synchrotron radiation mammography,‖ Physics in Medicine and Biology, vol. 45, no. 12, pp. 3509–3523.
[2] Graffigna J.P., and Romo R., "Imágenes en Medicina-UNSJ", Fundamento de Imágenes Médicas, 2003.
[3] Castelli E., Dreossi D., Rigon L., Abram A., Arfelli F., P. Bregant, et al., "The mammography project at the SYRMEP beamline" European Journal of Radiology, vol. 68, pp. S58-S62, 2008.
[4] Arfelli F., Castelli E., Rigon L., Bonvicini V., Bravin A., Cantatore G., et al., "Mammography with Synchrotron Radiation: Phase-Detection Techniques1" Radiology, vol. 215, pp. 286-293, 2000.
[5] Castelli E., Tonutti M., Rigon L., Arfelli F., Longo R., Quaia E., et al., "Mammography with synchrotron radiation: first clinical experience with phasedetection technique" Radiology, vol. 259, pp. 684-694, 2011.
[6] Ruiz Y., “Initial steps of objective measurements of image quality in phase-contrast vs. digital mammography”. CLAIB _2014
[7] Ruiz Y. Pérez M., Martínez D., Barreto M., Fleitas I., Mora R., RigonL., Tromba G. and Bregant P. et al. “Objective measurements of image quality in synchrotron radiation phase-contrast imaging versus digital mammograph”.International Journal of ComputerAssistedRadiology and Surgery. 2014.
[8] Ruiz Y ., Perez M., Diaz R. “ Mediciones de calidad de imagen en mamografía con contraste de fase mediante haces de sincrotrón vs. mamografía digital”. ConferenciaInternacional de Procesamiento de la Información.2017
[9] Tomasi C., Manduchi P.,“ Bilateral Filtering for Gray and Color Images,” Proceedings of the IEEE International Conference on Computer Vision, Bombay, India, 1998.
[10] “Image Processing Toolbox™ User’s Guide ”Matlab 2011a versión 7.8 pp-11
[11] Young I., Gerbrands J., “Fundamentals of Image Processing” pp-56-93
[12] “Image Processing Toolbox™ User’s Guide ”Matlab 2015a .
[13] Taswell C. “The what, how, and why of wavelets shrinkage denoising”. ComputSciEng. 2000;1: 12---9.22.
[14] Pérez M., Lorenzo J., Pérez J., Falcón A., “Filtrado wavelet en estudios planares óseos de Medicina Nuclear” Imagen Diagn. 2017; 8(1):3---14.
[15] Alvarez M., Oliveira M., Pina M., “Comparison between subjective and quantitative methods for assessing the resolution limit of radiographic systems” RevistaBrasileira de Física Médica.2012; 6(1):13-7.