Desarrollo de un fantoma clínico mediante impresión 3D para control de calidad en tomografía computarizada

Abstract

El presente trabajo, Desarrollo de un fantoma clínico mediante impresión 3D, surge del interés por explorar el potencial de la impresión 3D en el área de la física médica. Esta tecnología llamó mi atención a mediados de la carrera, debido a la libertad creativa que propone y la posibilidad de materializar ideas como la fabricación de equipos de laboratorio o dispositivos que permitan realizar mediciones en el ámbito de la física. Durante el desarrollo de este proyecto fue necesario abordar temas como el análisis de imágenes médicas, la caracterización de parámetros físicos de calidad y el uso de software especializado para la evaluación de imágenes. Conforme investigaba cada uno de estos temas, mi interés por ellos aumentó aún más. Esta tesis representa mi aporte, desde un enfoque accesible, a la mejora del control de calidad en equipos de imagenología médica.

In this research work, the development of a clinical phantom using 3D printing through the fused deposition modeling (FDM) technique is presented, with the objective of evaluating basic quality control parameters in computed tomography. The phantom design simulates the dimensions of a head module and incorporates inserts for performing spatial resolution, uniformity, contrast, and noise tests. Additionally, in a separate section, a central and peripheral cavity was incorporated for dosimetric measurements using a RADCAL 10X6-3CT ionization chamber.

Image acquisitions were performed using a Canon Aquilion Lightning CT scanner, applying a standard clinical protocol for head studies. The images were exported and analyzed using the software RadiAnt DICOM Viewer and ImageJ, where regions of interest (ROIs) were used to quantify the quality parameters.

The results obtained remained within acceptable ranges according to the consulted literature and the TEC-DOC 1958 standard. Regarding the uniformity of the printed material, good homogeneity was observed in the PETG, with an internal variation no greater than 5 Hounsfield Units (HU) between the measured regions, confirming a uniform distribution of the material during the printing process. Furthermore, the uniformity analysis in a water section showed a maximum difference of 2 HU, meeting the criteria established for this test.

In the low-contrast test, the cylinders with larger diameters and smaller thicknesses achieved a signal-to-noise ratio greater than 30, while the smaller cylinders presented minimum values of approximately 8.7. Despite these differences, all cylinders were visible, demonstrating that the insert design allows for an adequate evaluation of the system’s sensitivity to low-contrast objects.

Regarding spatial resolution, it was determined that the maximum resolution limit achieved was 0.7 lp/mm, mainly conditioned by the nozzle diameter used in the 3D printing process. In addition, the preliminary CTDI100 test demonstrated the phantom’s compatibility for performing dosimetric measurements.

These results lead to the conclusion that 3D printing technology is a viable alternative for the development of useful phantoms in quality control processes for medical imaging.