The results identify promising design criteria for extending atomic force microscopy, or AFM, beyond its familiar roles in imaging and local material characterization. AFM indentation works by lowering a flexible cantilever into a sample and recording how the probe deflects in response to contact. A laser reflected from the cantilever is directed onto a position-sensitive photodiode, which monitors and records the cantilever deflection.

The AFM output was subsequently converted into force–indentation curves, which researchers fitted with a contact model to estimate properties such as Young’s modulus. That modulus represents a material’s intrinsic elasticity, whereas apparent stiffness is the slope of the force–indentation curve and therefore also incorporates the mechanical environment surrounding the specimen. That distinction matters because a rigid support beneath a soft specimen resists the same indentation and can make the material appear stiffer than it is.

The effect is commonly treated as an artifact in conventional AFM, but the study deliberately used it as a signal. In earlier work, this substrate effect allowed the turgor pressure of bacterial and plant cells to be inferred from changes in apparent stiffness. The new experiment asked whether a related response could be detected in a larger tissue-like structure and, more specifically, whether it could reveal pressure transmitted through a corneal surrogate.

The artificial cornea was a homogeneous proprietary material with a Young’s modulus of approximately 25 kilopascals, a value the investigators said was comparable to their earlier measurements on cadaveric corneas. It was mounted in a custom sealed-end pressure chamber and indented at its apex with a custom AFM system driven vertically at 15 micrometres per second. The pressure transducer covered 0 to 0.1 bar, or 0 to 75 mmHg, with a stated static accuracy of ±0.1% and a total error band of ±0.25%.

Representative recordings were within approximately ±0.3 mmHg of the target pressures. At least ten repeated measurements were acquired at each of the three pressure levels, with indentation remaining below 2.50 micrometres. Only approach data were analysed.

The initial nonlinear region of each force–indentation curve was fitted with a Hertzian model for a spherical indenter, using an assumed Poisson’s ratio of 0.49, while a manually selected linear region was used to calculate apparent stiffness. Fourteen commercially obtained tipless probes were used first to examine working length and spring constant as separate variables. Their working lengths ranged from 90 to 130 micrometres, and their spring constants were measured before and after any modification using thermal power spectral density analysis.

Eight additional probes were fitted with epoxy-mounted glass beads measuring 50 to 100 micrometres in diameter, creating spherical indenters whose radii were determined from microscope images. Among the tipless probes, spring constant had a marked influence on both measured outcomes. At 15 mmHg, spring constant correlated significantly with apparent stiffness (p less than 0.001) and with Young’s modulus (p less than 0.05).

Adding spring constant to statistical models that already included working length raised the explained variance from 0.01–0.07 to 0.425–0.542 for apparent stiffness and from 0.001–0.034 to 0.427–0.476 for Young’s modulus. All likelihood-ratio comparisons were significant at p less than 0.001. By comparison, working length alone was a weak predictor.

One 130-micrometre probe nevertheless produced an estimated modulus of 508.0 kPa at 35 mmHg, with a 95% confidence interval of 416.3–599.8 kPa, versus 327.5 kPa for the 90-micrometre probe and 321.7 kPa for the 110-micrometre probe. Geometry became more consequential after the spherical indenters were added. Across modified probes with medium and high spring constants, the ratio of working length to indenter radius was positively associated with apparent stiffness and modulus.

This did not hold for very low spring constants, which the authors attributed to an inability to generate enough force for adequate indentation, leaving measurements dominated by the instrument’s threshold rather than tissue mechanics. A modified probe with a spring constant of 36.39 N/m, a 34-micrometre indenter radius and a 40-micrometre working length gave the most robust response. Its working-length-to-radius ratio was 1.2, the lowest in the high-spring-constant group.

Yet a second probe with a similar spring constant of 35.36 N/m did not behave equivalently, demonstrating that high stiffness alone was insufficient to guarantee reliable pressure detection. Adding a spherical bead also changed how the measurements responded to pressure. At spring constants around 8.2 N/m, tipless probes produced higher apparent stiffness and modulus estimates than modified probes, whereas the difference was not evident near the lower end of the tested spring-constant range.

The authors proposed that the bead shortened the effective working length and restricted bending, while its expanding contact area produced nonlinear Hertzian mechanics concentrated near the center of contact. A tipless probe instead acted more like a flat punch over a broader region. They therefore proposed that tipless designs may better capture bulk corneal behavior, whereas spherical tips may emphasize localized compression, although this distinction requires further testing.

The findings remain preliminary because the pressure-sensitive response was demonstrated in a manufactured, homogeneous surrogate rather than living or postmortem human corneal tissue. Native corneas contain collagen architecture, cell interactions, hydration gradients and regional variations that the artificial material does not reproduce. The investigators also applied a spherical-contact model to tipless probes, manually selected the transition between nonlinear and linear curve regions, assumed a Poisson’s ratio and tested spring constants no higher than 37 N/m.

Visual positioning at the corneal apex also introduced variability, and measurements away from that site may be needed to avoid artifacts caused by membrane displacement. The study therefore demonstrates a design strategy in a controlled model rather than a validated method for measuring intraocular pressure in patients. The long-term objective is to use pressure-dependent stiffness changes in the cornea as a possible route to intraocular-pressure measurement, including pressure differences on the order of 1 mmHg that may be relevant to glaucoma monitoring.

It also did not test healthy people or animals, and it did not show that AFM could replace established tonometry. What the experiments do show is that the pressure signal depends strongly on how the probe is built: a high spring constant can improve coupling to a pressurized substrate, but effective working length and contact geometry may determine whether that signal remains measurable across the pressure range. Confirming those criteria in postmortem corneas will be the next essential step before the method can be evaluated in more biologically realistic tissue.

Subject of Research: Experimental optimization of AFM cantilevers for detecting pressure beneath an artificial cornea

Article Title: Optimization of Atomic Force Microscopy Cantilever Specifications for Pressure Measurement in Biological Tissues

Article References: deHaseth, N. H., Penarredonda, A., Orlando, G., Infante, M., Rodriguez, O. G., Moy, V. T., Knight, O. J., & Ziebarth, N. M. (2026). Optimization of Atomic Force Microscopy Cantilever Specifications for Pressure Measurement in Biological Tissues. Annals of Biomedical Engineering. https://doi.org/10.1007/s10439-026-04420-2

Image Credits: AI Generated

DOI: 10.1007/s10439-026-04420-2

Keywords: atomic force microscopy, cantilever design, artificial cornea, pressure sensing, intraocular pressure, corneal biomechanics, apparent stiffness, substrate effect, nanoindentation, spring constant, Optimization, Atomic