Poster Z11 - Computational Modeling of Mechanical Strain in Brain Tissue Induced by Microelectrode Array Implantation

Introduction: Implantable microelectrode arrays (MEAs) play an essential role in the chronic recording of cortical neurons and the development of brain-machine interfaces. However, chronic brain implants face challenges such as signal loss, a low SNR, and unstable recordings over time. A significant issue limiting the longevity of MEAs is the immune response triggered at the implant site [1]. One of the major cata lysts for the immune response at the probe-tissue interface is the mechanical strain induced in the brain tissue due to the probe’s geometry (length, width, and thickness). Above, difficulties have prompted scientific investigations to thoroughly assess the biomechanics of MEA implants, specifically investigating the effect of mechanical strain on neural tissue, at the site of device implantation. The measurement of these mechanical strains is challenging since they are located within the brain tissue. This study investigates the development of a computational model to assess the mechanical strain induced in brain tissue by implanting different geometries of MEAs. This model also addresses the effects of various neural probe displacements on brain tissue and the associated strain patterns

Materials and

Methods: To investigate tissue strain caused by displacement of two distinct MEA geometries, we developed a three-dimensional (3D) computational model using COMSOL Multiphysics 6.1. This model comprises two components: a silicon-based single shank MEA and surrounding brain tissue unit. The geometry of the brain tissue block is 20 mm in length, 20 mm in width, and 20 mm in height, ensuring that the boundary effects are negligible due to the dimensions being ten times larger than the probe's cross-sectional area. The brain tissue, primarily composed of gray matter, was modeled as a linearly elastic material with isotropic properties. For a silicon-based single shank MEA, we constructed two different MEA geometries for the simulation: MEA-1, with dimensions of 1.125 mm in length, 0.125 mm in width, and 0.025 mm in thickness; and MEA-2, measuring 10 mm in length, 0.07 mm in width, and 0.02 mm in thickness. The walls of the brain tissue were fixed in all directions to act as model boundary conditions. To replicate the relative micromovement between the skull and brain tissue we applied displacements ranging from 1-5 μm, to model vascular micromotion, at a range of 10-100 μm for respiratory micromotion on the top surface of the probe. An extra fine mesh was employed to enhance the accuracy of the model. Finally, the simulations for each MEA geometry were performed separately, and the output metrics included von Mises stress, volumetric strain, and equivalent deviatoric strain near the probe tip surface.

Results, Conclusions, and Discussions:

Results:
Our model efficiently simulated the tissue strain profiles associated with two distinct MEA geometries. In both configurations, the highest strain was observed at the interaction point of the probe tip, with high strain concentrations also detected around the probe's sharp corners. Volumetric strains were measured as 0.113 for MEA-1 and 0.00713 for MEA-2, respectively. Notably, the equivalent deviatoric strain values showed an exponential decline with increased distance from the tip intersection in the brain tissue. Our analysis further revealed that MEA-1 consistently experienced higher strain across all applied displacements compared to MEA-2, suggesting that MEA-2 may exhibit better mechanical stability.

Conclusion:
The tissue strain results from our models show significant differences between the two MEA geometries investigated. This comparative analysis reveals information on how each shape may affect the immune response and long-term stability, which are critical aspects for chronic brain recording applications. Our findings suggest that MEA-2 is more suited for reducing tissue disruption over time and improving implant performance. Nevertheless, further research is necessary to corroborate these simulations and expand upon the findings with a wider variety of metrics. Such studies are necessary to understand the impact of MEA design on neural tissue immunological response, ultimately increasing the efficacy and sustainability of brain-machine interfaces.

Acknowledgements (Optional): [1] Al Abed, J. Amatoury, and M. Khraiche, “Finite Element Modeling of Magnitude and Location of Brain Micromotion Induced Strain for Intracortical Implants,” Frontiers in Neuroscience, vol. 15, 2022, Accessed: Nov. 17, 2023. [Online]. Available: https://www.frontiersin.org/articles/10.3389/fnins.2021.727715
[2] C. Cointe et al., “Scalable batch fabrication of ultrathin flexible neural probes using a bioresorbable silk layer,” Microsyst Nanoeng, vol. 8, p. 21, 2022, doi: 10.1038/s41378-022-00353-7.
[3] J. Subbaroyan, D. C. Martin, and D. R. Kipke, “A finite-element model of the mechanical effects of implantable microelectrodes in the cerebral cortex,” Journal of Neural Engineering, vol. 2, no. 4, p. 103, Oct. 2005, doi: 10.1088/1741-2560/2/4/006.
[4] Z. Xiang, J. Liu, and C. Lee, “A flexible three-dimensional electrode mesh: An enabling technology for wireless brain-computer interface prostheses,” Microsyst Nanoeng, vol. 2, p. 16012, 2016, doi: 10.1038/micronano.2016.12.
[5] C. Marin and E. Fernández, “Biocompatibility of intracortical microelectrodes: current status and future prospects,” Front Neuroeng, vol. 3, p. 8, 2010, doi: 10.3389/fneng.2010.00008.