2281 - Creation of Novel Inexpensive Low Artifact Brachytherapy Contrast Markers Compatible with CT and MRI

M. Farris¹, C. Lackey², N. Razavian², M. Lipford³, A. C. Snavely⁴, S. E. Glynn¹, T. G. Marshall², W. Dolesh², M. M. McKee², J. D. Ververs², M. Munley², R. T. Hughes², B. A. Frizzell², D. Brown², and P. J. Black²; ¹Department of Radiation Oncology, Wake Forest University School of Medicine, Winston-Salem, NC, ²Department of Radiation Oncology, Wake Forest University School of Medicine, Winston Salem, NC, ³Department of Radiology, Wake Forest School of Medicine, Winston Salem, NC, ⁴Department of Biostatistics and Data Science, Wake Forest University School of Medicine, Winston Salem, NC

Purpose/Objective(s): Current interstitial brachytherapy needles utilize radio-opaque markers that are expensive, largely MR incompatible, and/or create significant artifact on CT/MR imaging. Here, we developed a novel 3Dprinted radio-opaque marker composed of metal-infused polylactic acid (PLA) and evaluated its performance on CT and MR. Materials/

Methods: Novel radio-opaque markers were 3D-printed using PLA filament infused with either copper, bronze, or brass. Physical parameters were designed to match existing brachytherapy needles. Solid and hollow markers were designed, with hollow markers capable of containing viscous contrast media. Markers were inserted into a tissue-equivalent prostate phantom and images were acquired using CT and MR (1.5 and 3.0 T) techniques. Commercially available stainless steel and nitinol markers were imaged with the same modalities. Imaging artifact was quantified using ImageJ and a treatment planning system. CT metal artifact (CTMA) was quantified by computing standard deviation in the volume around each marker. Contrast-to-noise ratios (CNR) were evaluated for both imaging modalities, demonstrating better marker visualization compared with commercially available. Artifact differences between each set of markers were compared using F-tests and CNRs were compared using bootstrap tests, with a P-value <0.05 considered significant.
Results: The average CNR for copper, brass, and bronze PLA markers were 33.6, 33.9, and 34.0, respectively, while the average CNR for the commercially available nitinol and wire markers was 28.1, and 7.7, respectively (P<0.001 for all pairwise comparisons). CTMA for copper, brass, and bronze PLA markers were 29.9 HU, 31.4 HU, and 34.5 HU, respectively, while the average artifact for the nitinol and wire markers was 42.7 and 193.5 HU, respectively (P<0.001).
Conclusion: When compared with commercially available steel or nitinol, 3D-printed markers have enhanced visibility, reduced imaging artifact, and superior CT- and MR contrast. These novel markers can help improve the accuracy of interstitial brachytherapy implants for treatment planning and facilitate MR only planning. Abstract 2281 - Table 1: Measurement of CT metal artifact using standard deviation of the Hounsfield units within the region of interest. Pairwise P values are provided for F tests comparing markers at each intersection

Standard Deviation P-values (F-Test)	Nitinol 42.7 HU	Wire 193.5 HU	Air 28.0 HU
Hollow Copper PLA 31.9 HU	<0.001	<0.001	<0.001
Solid Copper PLA 29.9 HU	<0.001	<0.001	<0.001
Solid Brass PLA 31.4 HU	<0.001	<0.001	<0.001
Solid Bronze PLA 34.5 HU	<0.001	<0.001	<0.001