Design and fabrication of 3D-printed patient-specific soft tissue and bone phantoms for CT imaging

The objective of this study is to create patient-specific phantoms for computed tomography (CT) that have realistic image texture and densities, which are critical in evaluating CT performance in clinical settings. The study builds upon a previously presented 3D printing method (PixelPrint) by incorporating soft tissue and bone structures. We converted patient DICOM images directly into 3D printer instructions using PixelPrint and utilized stone-based filament to increase Hounsfield unit (HU) range. Density was modeled by controlling printing speed according to volumetric filament ratio to emulate attenuation profiles. We designed micro-CT phantoms to demonstrate the reproducibility and to determine mapping between filament ratios and HU values on clinical CT systems. Patient phantoms based on clinical cervical spine and knee examinations were manufactured and scanned with a clinical spectral CT scanner. The CT images of the patient-based phantom closely resembled original CT images in texture and contrast. Measured differences between patient and phantom were less than 15 HU for soft tissue and bone marrow. The stone-based filament accurately represented bony tissue structures across different X-ray energies, as measured by spectral CT. In conclusion, this study demonstrated the possibility of extending 3D-printed patient-based phantoms to soft tissue and bone structures while maintaining accurate organ geometry, image texture, and attenuation profiles.


Introduction
In computed tomography (CT) research and clinical practice, anthropomorphic and geometric phantoms play a crucial role.Highly accurate, customizable, and realistic phantoms are particularly valuable for a variety of purposes, including maintenance, optimization, and development of software and hardware components of scanners.In recent years, there have been signi cant advancements in three-dimensional (3D) printing technology, resulting in numerous studies on 3D-printed patient-based phantoms for medical imaging [1]- [5].Compared to conventional phantoms, 3D-printed phantoms are highly accessible, customizable, and cost-effective.For example, inexpensive and widely available fused deposition modeling (FDM) printers can create high-quality anthropomorphic phantoms that accurately depict human anatomy at reasonable costs.
Conventional 3D printing techniques prioritize the replication of object and organ shapes.Typically, these approaches include segmenting organs of interest from CT scans according to their speci c densities (HU), converting the results into surface meshes (STL les), 3D-printing each object separately, and then assembling them into a complete phantom [3], [4], [6], [7].However, each 3D-printed component has a uniform Houns eld unit (HU), resulting in phantoms with lacking realistic image textures because their HUs cannot be modulated pixel-by-pixel [8]- [11].Furthermore, the lack of natural transitions between different regions, e.g., organs, leads to loss of detail.A promising alternative is to directly translate digital imaging and communications in medicine (DICOM) image data into G-code.G-code is a Computer Numerical Control (CNC) programming language.G-code instructions tell the printer to move in speci c directions and at speci c speeds to produce a speci c shape or object.One means of controlling the density (as required for CT phantoms) is to vary the lament extrusion rate (per unit time) on a pixel-by-pixel basis while maintaining a constant printing speed.A similar approach was used by Okkalidis et al. [12]- [17]. in conjunction with edge detection and morphological operations to enhance and separate organs.Such processes still yield segmentation errors and loss of small features.Altering the line width by varying the extrusion rate alone does not provide su cient spatial resolution due to the inherently slow response time of the extrusion process.Our group recently developed PixelPrint [18], a methodology that combines a software tool as well as a standard FDM printer to create phantoms [19]- [23].In PixelPrint, DICOM images of the original patient are directly converted into G-code on a pixel-by-pixel basis.In order to emulate attenuation at each voxel, density is modeled as a ratio of lament to voxel volume, generating partial volume effects.The lament ratio is continuously modi ed by varying the printing speed.Polylactic acid (PLA), a common printing lament, allows a print range approximately from − 850 to 200 HU at different lament ratios, and has been used to print various patient-based lung phantoms [19].
In parallel, signi cant progress has been made in developing lament materials suitable for FDM printing in medical applications.Several studies have explored and compared different types of lament materials for printing human soft tissue and bones [24]- [27].Conventional materials, such as PLA and acrylonitrile butadiene styrene (ABS), are widely available and easy to print with.They have densities ranging from 0.8 to 1.2 g/ml and can represent various human soft tissues for CT or X-ray examinations.
Special materials, such as thermoplastic polyurethanes (TPU), can provide distinct physical properties to the print, i.e. durability, strength, and elasticity.Speci cally for bone, materials tailored for clinical applications have been introduced for 3D-printed implants.They are biodegradable by the patient's osteoclasts.As a result, printed objects with such materials can be fused with the patient's bone, through remodeling during the osteo-cycle [28]- [30].Additionally, denser PLA lament mixed with gravimetric powdered stone (PLA/Stone) has become commercially available.In previous studies, this type of lament has been utilized for printing phantoms for both diagnostic imaging and radiation therapy [8], [14], [31]- [35].For printing even higher density objects, commercially available lament materials mixed with micro metal powders, i.e. iron or copper, have also been utilized in phantom studies [6].
This study optimized several aspects of the previously published PixelPrint technique, including lament line spacing and print speed.Furthermore, StoneFil lament, a type of PLA/Stone lament, was utilized to expand the density range of our phantoms in order to print bony structures.Our results illustrate that the PixelPrint technique can create realistic phantoms of the human spine and knee joint with surrounding soft tissue.The resulting phantoms achieved accurate geometry, image texture, and attenuation.Moreover, the presented phantoms exhibited similar spectral attenuation pro les to that of bone structures, which enables their use in various spectral CT applications.
Materials and Methods

PixelPrint and 3D printing
The previously published PixelPrint algorithm was used to create G-code from CT image data to produce 3D-printed phantoms [18].Brie y, density information was extracted from the clinical patient images to generate lament lines that varied in width according to the HU of individual pixels.These lines were uniformly spaced within each layer and perpendicular on adjacent layers.By adjusting the lament line widths pixel-by-pixel, volumetric lament per unit space, or in ll ratio, was varied despite only using one type of lament.These different in ll ratios then produced different attenuation in CT images due to the partial volume effect.
In this study, the lament lines were equally spaced at 0.5 mm.The width of the lament line changed at resolution of 0.167 mm.The minimum and maximum line widths were 0.2 and 0.5 mm, corresponding to the in ll ratio ranging between 40% and 100%, respectively.Keeping a constant extrusion rate, the print head traveled at varying speeds based on the width of the extruded lament line.The slowest speed was 180 mm/min for the widest width of 0.5 mm, while the fastest was 450 mm/min for the smallest width of 0.2 mm.Each layer had a uniform height of 0.2 mm.The resulting volumetric rate of lament extrusion during the whole print remained constant at 18 mm 3 /min.To prevent overlapping of lines in consecutive layers with the same lament line direction, an offset of 0.167 mm (1/3 of the 0.5 mm line spacing) was introduced.
All phantoms were printed with Lulzbot TAZ 6 or Sidekick 747 (Fargo Additive Manufacturing Equipment 3D, LLC Fargo, ND, USA), paired with M175 v2 tool heads and 0.40 mm steel nozzles.StoneFil lament (FormFutura, AM Nijmegen, the Netherlands) with a diameter 1.75 mm was utilized.The temperature of the nozzle was set at 200°C and the bed was warmed to 50°C to enhance adherence.Acceleration of the print head was to 500 mm/s 2 and the threshold (jerk setting) was 8 mm/s.

Phantom design
Micro-CT phantom.Three cylindrical phantoms were designed and produced using PixelPrint lament lines to examine their stability and reproducibility.These lament lines constructed a matrix smaller than the typical resolution limit of clinical CT scanners.Three phantoms were printed with identical G-code instructions.These phantoms are 60 mm in length and 20 mm in diameter.Each of them consists of four sections with different but homogeneous in ll ratios (100%, 70%, 50% and 30%).StoneFil lament lines were printed at a spacing of 1 mm in all four sections but with corresponding line widths of 1.0, 0.7, 0.5, and 0.3 mm, respectively.A thin outer layer was added to the phantom for support, particularly for low in ll ratio sections.
Calibration phantom.To compute the conversion between StoneFil lament in ll ratios and HUs, a calibration phantom was designed.The phantom is a cylinder with a diameter of 10 cm and height of 1 cm.It consists of seven equally divided pie slice-shaped sections.Each section was printed at a xed line spacing of 0.5 mm but with different lament line widths (0.2-0.5 mm), corresponding to seven in ll ratios (40-100%, with 10% intervals).
Cervical vertebrae phantom.Institutional Review Board (IRB) of University of Pennsylvania approved this retrospective study.Informed consent was obtained from all subjects and their legal guardians.All methods were performed in accordance with relevant guidelines and regulations.A cervical vertebrae phantom was created based on a patient image volume (10 x 10 x 10 cm 3 ) that was acquired on a clinical CT scanner (Siemens SOMATOM De nition Edge, Siemens Healthcare GmbH, Erlangen, Germany) at a tube voltage of 120 kVp with a standard diagnostic protocol.Table 1 lists detailed acquisition and reconstruction parameters for the patient scan.The patient data consist of four cervical vertebrae (C4 to C7), including the trachea and esophagus.A circular region of interest with a diameter of 10 cm was cropped in axial slices to form the phantom.HUs were converted to in ll ratios based on the calibration phantom.
Knee phantom.A knee phantom was similarly generated using a patient scan on a clinical dual-layer CT scanner (IQon spectral CT, Philips Healthcare, the Netherlands) at a tube voltage of 120 kVp, as detailed in Table 1.The data were publicly available from Philips Healthcare.A circular region of interest with a diameter of 10 cm was cropped from the axial slices of the patient's left knee.HUs were then converted to in ll ratios.

Data acquisition
Three micro-CT phantoms were separately scanned on a commercial micro-CT (U-CT system, MILabs, CD Houten, the Netherlands) with a tube voltage of 50 kVp.In addition, these phantoms were also scanned on a clinical dual-layer CT system (IQon spectral CT, Philips Healthcare, the Netherlands) at a tube voltage of 120 kVp with a high-resolution protocol and a small eld-of-view of 100 mm.Additional acquisition and reconstruction parameters of the two scans are listed in Table 2. Micro-CT images were exported from the scanner and reprocessed with a multi-planar reconstruction algorithm (MPR) in Horos (Horos Project, Annapolis, MD, USA) to ensure lament lines were parallel to the axial plane.
The calibration, cervical vertebrae, and the knee phantom were scanned inside the QRM chest phantom (Quality Assurance in Radiology and Medicine GmbH, Möhrendorf, Germany) with the clinical dual-layer CT system.Protocol parameters matched those of the original clinical examination of the patient, with the same pixel spacing and slice thickness in Table 1.For the cervical phantom, a 400 mg/ml QRM hydroxyapatite (HA) insert was additionally scanned with the phantom as a reference for bone mineral density.For both patient-based phantoms, additional high dose scans were performed at 1000 mAs while keeping the other scanning parameters the same.This high exposure scan was included to reduce noise for image quality comparisons.

Calibration and data analysis
For computing the conversion between HUs and in ll ratios, mean and standard deviation HU values of seven areas were measured in the calibration phantom.Square regions of interest (ROI) of 19 x 19 pixel 2 (13 x 13 mm 2 ) were manually placed in each of the seven density regions within 10-mm-thick center of the phantom.A linear regression was computed, and the resulting Pearson's correlation coe cient (r) was reported.All measurements were performed on a workstation with ImageJ (U. S. National Institutes of Health, https://imagej.nih.gov), and all analyses were computed with Python (Python Software Foundation, https://www.python.org/).
For the cervical vertebrae phantom and the knee phantom, CT images were exported from the scanner and registered to the original patient data (2D-wise) using the OpenCV Library (Open Source Computer Vision Library [36], https://opencv.org).Mean and standard deviation in regions of interest for different tissue types were measured.Line pro les of the phantom scan were also compared with the original patient scan.Additionally, virtual monoenergetic images from 40 to 200 keV were extracted to quantify the spectral response of the bone regions within the patient-based phantoms.

Results
The high reproducibility of PixelPrint was demonstrated by comparing three identically manufactured phantoms (Fig. 1).In micro-CT scans of the phantoms, the grid-like structures generated by PixelPrint were clearly visible.Filament lines printed within each region had equal spacings of 1 mm and a constant width in all three phantoms in the micro-CT scans.The layered structure with introduced offsets (1/3 of 1 mm line spacing) was distinctly visible in orthogonal views (Fig. 1f, 1g, 1h).However, in clinical CT scans with high resolution protocols, these structures were imperceptible because their size was smaller than the detector resolution.Instead, they appeared as constant regions due to partial volume effect (Fig. 1e).Furthermore, both the micro-CT and clinical CT scans showed a high linear relationship between in ll ratios and mean HUs in four regions (Pearson's correlation coe cient r = 0.984 and 0.982, respectively).
In the calibration phantom, the in ll ratio and HU also demonstrated excellent linearity across the seven regions (Fig. 2).The highest in ll ratio (100%) region measured 851 ± 24.7 HU, while the lowest in ll ratio (40%) measured − 227 ± 25.4 HU.Pearson's correlation coe cient of greater than 0.99 indicated a very high positive linear correlation between in ll ratios and HUs.A conversion equation was computed for converting HU to in ll ratio: (%) Patient phantoms showed high accuracy.Line pro les indicated a match in HUs between the CT image of the cervical vertebrae phantom and the patient data (Fig. 5).Quantitative measurements in selected regions of trabecular and cortical bones, as well as adipose-and muscle-like soft tissues, are provided in Table 3. Measurements indicated that, except for the cortical bone, all other regions had differences of less than 15 HU compared to the patient image.Due to the density limitations of the utilized lament, HUs for the cortical bone (region 3 in Fig. 6) were lower than expected.
Comparable spectral characteristics of the phantom to those of human bone were observed.Figure 6 depicts the spectral attenuation pro le of various regions of interest (marked in the left panel) and a 400 mg/ml hydroxyapatite insert (displayed in dark blue in right panel).It is noteworthy that the phantom was fabricated using only one type of lament, and thus, the background, which represents soft tissue, has arti cial amounts of calcium.

Discussion
This paper demonstrated how PixelPrint can be utilized to create patient-speci c 3D printed bone and soft tissue CT phantoms using one lament.Our approach provides economical and e cient means of producing high resolution CT phantoms, exhibiting excellent accuracy in HU and image texture characteristics in CT scans.These phantoms are useful for a wide range of academic research and clinical evaluation of CT performance.
In contrast to prior studies of image-based 3D printed bone phantoms using slices of the human head/skull [13], chest/thoracic cage [15], pelvis [14] and femoral shaft [6], this study printed the human cervical vertebrae with surrounding soft tissue.Human vertebrae particularly present a challenging task for 3D printing, as they contain intricate details and are comparatively smaller in size.Nevertheless, these areas, especially in combination with the adjacent tissues, are not only fundamental in clinical diagnostic applications, such as the assessment of severe fractures or degenerative diseases, but also crucial in surgical interventional planning.Our phantoms possess the potential to be utilized for those applications, such as optimizing CT protocols for the assessment of bone mineral density [37] among others.Here, only human cervical vertebrae and knee joint phantoms were printed, but the approach can be extended to any bone structure.With StoneFil lament, a range of approximately − 227 HU to 851 HU for a CT scans with a tube voltage of 120 kVp can be reliably printed using PixelPrint, with a deviation of less than 15 HU compared to patient data.This range covers most tissue types in the human body and is applicable to various research applications.
Continuing our previously published research on the PixelPrint lung phantom [18], [19], this study not only extended the types of human tissue printed, but also enhanced the resolution and stability of PixelPrint.Filament line spacing was reduced from 1.0 to 0.5 mm, potentially doubling the resolution capabilities of the printed phantoms.Phantoms produced using this approach can have greater lament coverage and ner details in a given area, serving as valuable tools to evaluate the e cacy of novel higher resolution CT systems such as photon-counting CT [38]- [40].Printing ner lines with PLA/Stone lament poses more challenges to printer stability control and requires ner system tuning.By optimizing extrusion rate, printing speed, nozzle temperature, and acceleration speed, PixelPrint can still produce highly accurate patient phantoms in reliable stability as demonstrated by qualitative and quantitative evaluation.Additionally, micro-CT acquisitions revealed that lament lines and underlying structure can be generated with high degree of consistency.
With the growing popularity and accessibility of 3D printing technology, a variety of printing laments are now available for printing human bone and soft tissue.Several studies have discussed materials for 3Dprinted phantoms in CT [24]- [26].Novel lament materials composed of hydroxyapatite and biocompatible, biodegradable polymers, such as CT-Bone (Xilloc Medical Int., Sittard-Geleen, the Netherlands), can be utilized for printing synthetic bone implants that rapidly induce bone regeneration and growth [41], [42].Filaments made from composites of fatty acids and ceramic powders have also been explored [28].However, bone-like laments available in the general market (FibreTuff, Toledo, OH, USA), suitable for medical surgery purposes [29], [30], do not necessarily have high radiometric densities and are not capable to reach much higher than 400 HU in CT scans.While cancellous bone is only about 300 to 400 HU in CT images, cortical bone can range from 500 HU and up to over 1900 HU [43].By contrast, materials such as vinyl and PLA with stone (PLA/stone) can offer up to nearly 1000 HU at 96.9% in ll ratio at tube voltage of 120 kVp, as they exhibit relatively higher X-ray absorption.Additionally, considering materials for spectral CT phantoms, high impact polystyrene (HIPS) based laments may be suitable for mimicking CT numbers in applications where energy dependence is important [26], because they show similar spectral pro les as the human body.In this study, we employed StoneFil lament, one type of PLA/stone lament.Unlike normal PLA, StoneFil lament is gravimetrically lled with 50% powdered stones, resulting in signi cantly higher material density and enabling denser printed objects.Carbonate calcium-containing limestones exhibit a similar X-ray response in CT to that of human bone, whose density can be attributed to hydroxyapatite.This property was re ected in the the spectral response of the printed vertebrae with its similarity to that of hydroxyapatite.
This study has a few limitations: (i) The lament used in our study did not encompass the entire range of Houns eld Units (HU) required for bone structures.Future research should focus on the development of next-generation laments that cover the full HU range while preserving spectral capabilities.(ii) The calcium-based material used in the printing process was applied to the entire print, including soft tissue regions.While this approach does not severely impact performance in conventional CT applications, it may have an in uence on the evaluation with spectral CT.To achieve the full dynamic range with spectral characterization for both soft tissue and bone, further development of multiple print head systems will be required.(iii) The printed phantoms were limited to a speci c eld of view.Future studies should explore the potential to print larger anatomical regions, such as the entire chest or abdomen.

Conclusion
Our study successfully showed the feasibility of using  All measurements are in HUs.Stdev stands for standard deviation.Patient and phantom images were assumed to be the same z location and registered 2D-wise.

Figure 3 .
Figure 3.Comparison between patient CT images and the PixelPrint cervical phantom images.Images in the rst row (a-d) are original DICOM images used to create the PixelPrint cervical phantom.Images on the second row (e-h) are the CT images of the phantom.All images have window level of 0 HU and width of 1200 HU.Sagittal and coronal images are not registered but are approximately at the same location.of the three different micro-CT phantoms scanned on a micro-CT.(e) Clinical CT image of one of the micro-CT phantoms.(f) -(i) Zoomed views of the regions enclosed by blue squares in (b) -(e).Window min/max are − 1000/2500 HU for micro-CT images and − 1000 /1000 HU for clinical CT images.

Figure 5 Line
Figure 5

Table 2 .
attenuation pro les for spectral CT, which can greatly bene t both academic research and clinical applications.Scan protocols the micro-CT phantom PixelPrint and stone-based lament to 3D-print patient-based bone phantoms with surrounding soft tissue for use in clinical CT applications.The resulting phantoms accurately replicated patient's CT imaging, including precise organ geometry, image texture, and Collimation width values are noted as single / total collimation width.

Table 3 .
Measured Houns eld units for different tissue types in patient and phantom.