Info
Background:
The incorporation of contrast agents into polymeric biomedical devices has been shown to improve the success rate for identifying device location and damage post-implantation [1]. However, bottlenecks for clinical adoption of longitudinal monitoring include access to imaging modalities and supporting the additional costs of periodic evaluation. Thus, imaging functionality incorporated into devices should be extended from traditional magnetic resonance imaging (MRI) and computed tomography (CT) to less costly imaging modalities that can be performed on-demand and preferably at the bedside, including magnetic particle imaging (MPI) and ultrasound.
Hypothesis/Objective:
By incorporating two types of nanoparticle contrast agents into 3D printed phantoms, we aim to demonstrate multimodal imaging of devices using MPI, micro-CT and ultrasound post-implantation.
Methods:
Standard curves of nanoparticle contrast agents were created using Modeling Clear V2 SLA based resin (ApplyLabWork, CA). For MPI contrast, iron oxide (Fe3O4, Synomag-D (104-00-501)) was used up to 5.5 mg/ml; for micro-CT, tantalum oxide (TaOx) nanoparticles were used, up to 72 mg/ml (160 mM tantalum). Phantoms were 3D-printed on a Form Labs 3 stereolithography (SLA) based printer. Nanoparticles were pre-mixed into the resin at 0.25 mg/ml Fe3O4 and 60 mg/ml TaOx and printed, followed by an IPA wash, and UV-cure. Nanoparticle content was confirmed using thermogravimetric analysis. Phantoms were imaged pre-implantation in phosphate buffered saline (PBS) and post-implantation in model tissue. MPI was conducted on a Magnetic Insight Momentum in 2D imaging mode with the 3D ROI tool to quantify signals, and micro-CT was performed with a Perkin-Elmer QuantumGX at 90keV, 88 µA and analyzed with Image J. Further, ultrasound was conducted at moderate brightness (moderate gain), high frequency (CHI mode), shallow image depth, and high contrast.
Results:
When considering multimodal device imaging, we utilized micro-CT, MPI, and ultrasound for their complementary strengths: MPI’s sensitivity, micro-CT’s resolution, and ultrasound’s convenience and portability. Further, the contrast agents chosen (Fe3O4 and TaOx) are known to be effective and biocompatible [2-3]. Standard curves confirmed that nanoparticle contrast was retained in resin, exhibiting a linear relationship with nanoparticle addition in both micro-CT and MPI. However, the amount of contrast required for a signal that could be differentiated from background was at least an order of magnitude higher in micro-CT than in MPI. Ultrasound didn’t require any contrast agent.
Porous phantoms were successfully 3D printed with nanoparticle addition across a range of feature sizes. It was found that nanoparticle settling could occur in resin over time, requiring homogenization of the resin prior to every print, or within 24 hours. A dual-contrast phantom with >8wt% total nanoparticle content (Fe3O4 and TaOx) was detectable in MPI, micro-CT and ultrasound, post-implantation in a model tissue. As shown previously, micro-CT was able to detect phantoms and their features with high resolution [4]. MPI 2D imaging was found to be informative for device presence and general location, but without extensive post-processing, the resolution did not allow for information on the device state and could be complicated by dual phantoms or the orientation of the implant in the 2D plane. Ultrasound imaging also confirmed the location of phantoms, but without details on the size or shape of the device.
Conclusion:
Utilizing nanoparticle contrast agents, phantoms incorporating multimodal imaging functionality could be 3D printed, an important step towards the transition to devices that are produced on-demand and are patient specific. Further, this study demonstrates trimodal visibility in MPI, micro-CT, and ultrasound capitalizing on the strengths of each modality. Overall, these results show that devices can be engineered to support real-time, non-invasive monitoring that can be more easily incorporated into the clinical workflow.
Image/Figure:
Click to view full size
Click to view full size
Image/Figure Caption:
Incorporating iron oxide (Fe3O4) and tantalum oxide (TaOx) nanoparticles into a 3D-printed resin phantom imparts contrast, allowing the phantom to be imaged via magnetic particle imaging (MPI), computed tomography (CT) and ultrasound, shown schematically in (a). Stereolithography printing could create phantoms with a range of feature sizes from (b) millimeter scale to (c) sub-millimeter, visualized via (1) micro-CT scan and (2) macroscopically. It was difficult to distinguish multiple devices placed in close proximity via (d) MPI imaging in 2D, and the intensity was affected by device placement, highlighted by repeated scans of two phantoms as they were moved apart: (1) schematic of phantoms, (2) MPI scans of phantoms and (3) corresponding maximum signal intensity with displaced distance. When implanted in a model tissue, phantoms with dual nanoparticles were visible in (e) MPI, (f) ultrasound and (g) micro-CT: (1,2) scan cross-sections, (3) 3D rendering of the phantom and bone; white arrow: without nanoparticles, orange arrow: nanoparticle incorporation. The HU scale is consistent across micro-CT images. Scale bars: (b, g) 5mm, (c) 1mm.
Author
Kendell Pawelec
Michigan State University