Master’s Thesis: Edmondson, David. Mechanical Performance of Additively Manufactured Soft Materials for Ultrasound and Biomechanical Simulation. Master’s thesis. Auburn University; 2025. Advisor: Schulze, Kyle D.
Repository record: Auburn University Electronic Theses and Dissertations
Can a multi-material printer produce organ-specific anatomy and ultrasound appearances directly in one build?
This thesis takes an intriguing first step toward that idea. Its point-of-care-ultrasound work combines a CT-to-printable-anatomy workflow with a screening experiment involving 16 formulations made on a Stratasys J5 Digital Anatomy Printer. The goal was to identify candidate materials for liver, renal-pelvis, and renal-medulla anatomy in a modular right-upper-quadrant trainer.
How the anatomical models were created
The researchers obtained 25 deidentified abdominal CT examinations and 28 deidentified MRI examinations from East Alabama Medical Center under an approved research process. They selected one CT dataset with favorable contrast for liver and kidney anatomy.
The modeling workflow included:
- 3D Slicer and TotalSegmentator for gross abdominal anatomy.
- Thresholding, smoothing, island removal, local thresholding, and manual editing for internal liver and kidney structures.
- Review with medical students during scan selection and segmentation refinement.
- Mesh repair and solid-body preparation in Meshmixer and Fusion 360.
- Substitution of several models—including the ribs, aorta, and inferior vena cava—where the image-derived models were considered inadequate.
The resulting digital assembly included detailed liver and kidney anatomy and surrounding RUQ structures. The complete assembly was prepared for printing but was not manufactured or tested as an integrated trainer in this thesis.
How the materials were tested
The researchers printed 16 rectangular coupons measuring 7.7 × 5.3 × 1.7 cm. Each coupon contained a different digital-material formulation made from:
- ElasticoClear, a rubber-like material;
- TissueMatrix, a soft tissue-oriented material;
- GelMatrix, a very soft gel-like material; and
- DraftWhite, a rigid resin used in small amounts to increase structure.
Every coupon also had an encapsulating outer layer made from 80% ElasticoClear and 20% TissueMatrix. The percentages therefore describe nominal printer-controlled digital-material recipes, not chemical mixtures prepared in a container.
The coupons were scanned with a Philips S4-1 phased-array transducer using the abdomen preset, 60% gain, 6 cm depth, and ultrasound gel. A medically trained professional experienced with RUQ ultrasound performed the scanning and categorized the images.
For each coupon image, the researchers analyzed a 1 × 0.5 cm region at a depth of 0.65 cm. They compared its mean grayscale intensity with regions selected from one deidentified human liver-and-kidney ultrasound examination acquired with the same probe and gain setting.
Principal findings
- TissueMatrix–GelMatrix blends S3 and S4 produced the brightest images and were judged to have high transmission.
- Increasing the contribution of ElasticoClear generally produced darker images, but those materials also attenuated the beam and obscured deeper features.
- DraftWhite offered some ability to adjust image intensity, although the tested formulations provided limited contrast and affected wave propagation.
- Print-layer lines were not visible within the analyzed ultrasound regions, suggesting relatively uniform internal coupon appearance at the tested scale.
- Changing digital-material proportions did not always produce a large or predictable change in ultrasound intensity. Composition, softness, brightness, and transmission did not move together as one simple variable.
- In a preliminary stacking experiment, changing the order of three coupons altered the appearance of the deeper layers. A material that looks promising alone can behave differently when another printed tissue is placed above it.
That last observation may be the most important result for complete manikin design. Independent coupon matching is insufficient; the entire acoustic path and every material interface must be evaluated together.
Why it matters for phantom builders
This thesis expands the role that 3D printing might play in ultrasound simulation. Instead of using a printer only for rigid anatomy, molds, or fixtures, a high-end multi-material system may be able to fabricate anatomically organized soft components with intentionally different ultrasound appearances.
That route is proprietary and likely more expensive than a cast gelatin, agar, or silicone trainer. It may still be valuable for small-volume, high-detail organ modules; repeatable pathologies; reference builds; or components that would be difficult to mold and assemble manually.