Abstract
Anatomically realistic organ replicas, or phantoms, are essential for accurate surgical studies and reproducible medical research. Replicating the human kidney requires mimicking its specific elastic properties. While paraffin and hydrogels are common, silicone offers superior shelf life and viscoelastic modularity. This study utilizes uniaxial compression testing and Cauchy stress modeling to evaluate commercial silicones. Results indicate that current off-the-shelf silicone mixtures fail to meet the dual target compressive moduli of 20 kPa (low strain) and 500 kPa (high strain), remaining significantly stiffer at low strain levels.
Introduction: The Need for Realistic Phantoms
To advance surgical training, pre-operative planning, and the validation of robotic systems, high-fidelity organ phantoms are required. While cadaveric tissue and animal models provide high anatomical accuracy, they are limited by high costs, ethical constraints, and inherent biological variability. Synthetic alternatives like hydrogels or paraffin are often used but suffer from rapid degradation and limited mechanical adjustability.
Silicone is a promising alternative due to its durability and tissue-like tactile properties. However, human soft tissue, particularly the kidney, exhibits complex "bilinear" elasticity. It is highly compliant under low strain (the "toe" region) and stiffens significantly under higher strain (the "heel" region). Replicating this behavior requires a material that can match two distinct targets: approximately 20 kPa at low strain and 500 kPa at high strain.
Methodology
The researchers developed a standardized pipeline to evaluate the mechanical suitability of various silicone mixtures:
- Uniaxial Compression (UC) Testing: Cylindrical samples of 00-10 ECOFLEX and M 4511 ELASTOSIL were prepared with varying percentages of silicone oil thinners (0% to 75%). These samples underwent compression testing at a fixed strain rate to determine their stress-strain profiles.
- Cauchy Stress Modeling: The nonlinear behavior was characterized using the Blatz model, a two-parameter mathematical framework used to define the constitutive relationship between stress and strain in hyperelastic materials.
- Segmented Linear Regression: To address the bilinear nature of the tissue, the team employed R-based segmented regression. This allowed for the identification of "change points" in the data, facilitating the calculation of two distinct elastic moduli: $E_1$ (low strain) and $E_2$ (high strain).
Results and Discussion
The study revealed a significant gap between commercial silicone performance and the mechanical requirements of a human kidney phantom. The mathematical models fit the experimental data exceptionally well, with $R^2$ values exceeding 98% for almost all tested mixtures.
The Bilinear Hurdle
The primary challenge identified was the "bilinear hurdle." While many mixtures could be tuned to approach the high-strain stiffness target, they remained far too stiff in the low-strain regime.
As shown in the ECOFLEX data, the 5% dilution mixture was the closest candidate. It achieved a high-strain modulus ($E_2$) of 470.61 kPa—near the 500 kPa target—but its low-strain modulus ($E_1$) was 61.81 kPa, more than three times the desired 20 kPa.
Convergence in ELASTOSIL
For the ELASTOSIL mixtures, the researchers observed a point of diminishing returns.
Beyond a 55% dilution, the material properties began to converge toward a specific limit (indicated by the red line in the figure). This suggests that simply increasing the amount of thinner is insufficient to achieve the extreme softness required for the "toe" region of kidney tissue.
Conclusion and Future Directions
The study concludes that off-the-shelf silicones, even when diluted, are currently inadequate for creating perfectly realistic kidney phantoms. While silicone remains a superior base material due to its longevity, custom material engineering is required to lower the low-strain modulus without sacrificing high-strain stiffness.
The researchers have made their full dataset and R modeling scripts (BlatzModel.R and Moduli.R) available via the Materials Cloud Archive. This open-access approach provides a benchmark for future work, allowing other researchers to use these mathematical frameworks to test new additives or alternative silicone formulations. Future efforts will likely focus on in silico modeling to predict phantom behavior in virtual surgical environments before physical production.
Summary updated