Optimización de la calidad de imagen en radiografía digital de la mama vs. dosis a la paciente
Archivos
Fecha
2020
Autores
Bernardo, Almeida António
Pérez Díaz, Marlen
Echevarría Torres, Waldo Jose
Díaz Amador, Roberto
Título de la revista
ISSN de la revista
Título del volumen
Editor
Resumen
A ́un no existe un conocimiento profundo de la calidad de imagenrespecto a la dosis, que se puede obtener empleando la tecnolog ́ıa“Computed Radiography” (CR). Hacia este objetivo se enfoc ́o elpresente trabajo. Se analizan im ́agenes de mamograf ́ıa digitalobtenidas de un maniqu ́ı antropom ́orfico. Se vari ́o la corrienteque alimenta al tubo de rayos X y el tiempo de adquisici ́on de laimagen y la energ ́ıa del haz de rayos X y se analiz ́o la influenciade esta variaci ́on sobre la dosis media glandular y la calidadde imagen. Para esto se emplearon las medidas matem ́aticasrelaci ́on se ̃nal-ruido, contraste imagen, relaci ́on contraste-ruido, ́ındice de similitud estructural, distancia espectral y error mediocuadr ́atico y se estimaron las dosis a partir de mediciones conuna c ́amara de ionizaci ́on. Tambi ́en se analiz ́o la calidad deimagen subjetivamente utilizando criterio de expertos. Se obtuvoun protocolo optimizado que reduce la dosis media glandularsignificativamente manteniendo buena la calidad de imagen.
There is still not in-depth knowledge of image quality with respect to the dose that can be obtained using “Computed Radiography” (CR). Toward this aim the present work was focused on. Digital mammography images obtained from an anthropomorphic phantom were analyzed. The current of the X-ray tube and the acquisition time as well as the energy of the X-ray beam were varied and the influence of these variations on the mean glandular dose and image quality were analyzed. Mathematical metrics like signal-to-noise-ratio, image contrast, contrast-to-noise ratio, structural similarity index, spectral distance and mean squared error were used to grade image quality. Subjective image quality analysis, using the criteria of expert observers, was also included and the doses were estimated from measures done with an ionization chamber. An optimized protocol was obtained which reduced the mean glandular dose significantly while a good image quality was maintained
There is still not in-depth knowledge of image quality with respect to the dose that can be obtained using “Computed Radiography” (CR). Toward this aim the present work was focused on. Digital mammography images obtained from an anthropomorphic phantom were analyzed. The current of the X-ray tube and the acquisition time as well as the energy of the X-ray beam were varied and the influence of these variations on the mean glandular dose and image quality were analyzed. Mathematical metrics like signal-to-noise-ratio, image contrast, contrast-to-noise ratio, structural similarity index, spectral distance and mean squared error were used to grade image quality. Subjective image quality analysis, using the criteria of expert observers, was also included and the doses were estimated from measures done with an ionization chamber. An optimized protocol was obtained which reduced the mean glandular dose significantly while a good image quality was maintained
Descripción
Palabras clave
Digital Mammography, Image Quality, Mean Glandular Dose, Optimization