In recent years, material extrusion-based additive manufacturing, particularly fused filament fabrication (FFF), has gained significant attention due to its versatility and cost-effectiveness in producing complex geometries. This paper presents the characterization of seven novel materials for FFF and twenty-two commercially available filaments in terms of X-ray computed tomography (CT) numbers, as tissue mimicking materials for the realization of 3D printed radiological phantoms. Two technical approaches, by 3D printing of cube samples and by producing cylinders of melted materials, are used for achieving this goal. Results showed that the CT numbers, given in Hounsfield unit (HU), of all the samples depended on the beam kilovoltage (kV). The CT numbers ranged from +411 HU to +3071 HU (at 80 kV), from −422 HU to +3071 HU (at 100 kV), and from −442 HU to +3070 HU (at 120 kV). Several commercial and custom-made filaments demonstrated suitability for substituting soft and hard human tissues, for realization of 3D printed phantoms with FFF in CT imaging. For breast imaging, an anthropomorphic phantom with two filaments could be fabricated using ABS-C (conductive acrylonitrile butadiene styrene) as a substitute for breast adipose tissue, and ASA-A (acrylic styrene acrylonitrile) for glandular breast tissue.

Conventional thermoluminescent radiation detectors (TLDs) are typically manufactured with fixed geometries—in this work, we attempt for the first time to employ additive manufacturing for the fabrication of customized radiation detectors, with the prospect of evolving geometrically complex and application-specific TLDs for incorporation into phantoms for use in medical radiology. While additive manufacturing has found applications in dose verification, quality assurance, imaging modalities, settings optimization of medical imaging devices, and the replication of soft and bone tissues, with a wide range of thermoplastic polymers used in extrusion printing having been extensively investigated for their radiological properties, there are no reported TLD examples, thus limiting the prospects of this technology. In this work, we manufacture composite polypropylene-CaSO₄-dysprosium filaments for 3D-printed TLDs, and we evaluate their response by irradiating 3D-printed prototypes in a medical linear accelerator. Our results validate their functionality within the limits imposed by the melting points of the polymeric binders, effectively qualifying them as TLDs for detection—but not dosimetry. Thus, 3D-printed TLDs incorporated in phantoms can enhance radiological investigations for detecting exposure to potentially harmful radiation.