• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 1
  • Tagged with
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

Anisotropic Mechanical Properties of the Guinea Pig Round Window Membrane

Wang, Wenbin January 2023 (has links)
Accessing the inner ear presents a significant challenge for the diagnosis and treatment of inner ear diseases. Many existing techniques to access the inner ear are invasive and can cause permanent damage to the cochlea. Recently, a novel microneedle has been fabricated to perforate the round window membrane (RWM) – a membrane sealing one of the two openings in the cochlea. These perforations enhance drug delivery into the inner ear, potentially improving the efficacy of therapeutics. Furthermore, they allow for the aspiration of perilymph samples, which is essential for diagnosing inner ear diseases. However, owing to limited knowledge about the mechanical properties of the RWM, certain technical aspects remain unexplored. Specifically, the interaction between the RWM and the microneedle during perforation is yet to be examined. This investigation is pivotal for the optimal design of microneedles — those robust enough to perforate RWMs yet delicate enough to minimize damage. In this thesis, we conduct a thorough examination of the guinea pig RWM, encompassing its geometry and its mechanical responses to pressures from the middle ear and inner ear. Additionally, we also formulate a comprehensive constitutive law for the guinea pig RWM. Our exploration begins with the creation of a U-Net model tailored to automatically segment the RWM. Despite the presence of other structures in the same image—such as bone, the basilar membrane, and ambient noise—the model proved invaluable for efficiently and automatically segmenting the RWM. To enhance accuracy, post-processing techniques like connected component analysis and majority voting were incorporated. Using this 3D model, we proceeded to study the RWM’s geometry. Recognizing the shrinkage observed in fixed RWMs, we integrated fresh RWM data to estimate the shrinkage ratio. Subsequently, we analyzed both the overall RWM thickness and that of the middle connective tissue layer—crucial metrics for future RWM modeling. Next, we proposed a method to evaluate the in-plane deformation of the RWM due to applied pressure. This involved using a bulge test system to pressurize and deform the RWM, combined with confocal microscopy to track stained nuclei or pre-introduced fluorescent beads on the RWM. We then utilized the coherent point drift (CPD) algorithm to measure the displacement of beads and nuclei. Results indicated that both markers could be successfully used to measure the RWM’s displacement. Further analysis revealed the in-plane Lagrangian strain of the RWM, with a significant observation being that the direction of maximum in-plane Lagrangian strain is perpendicular to the fiber direction. This underscores the crucial role of collagen fibers in determining the RWM’s mechanical properties. To conclude our study, we devised a constitutive law for the RWM, conceptualizing it as a combination of the ground substance and a family of dispersed fibers. This model was integrated into a FEBioStudio plugin, facilitating simulations of the RWM’s mechanical reactions to different pressures. Although our simulations closely aligned with experimental findings, some discrepancies were noted, likely stemming from an incomplete understanding of fiber dispersions. Nevertheless, our constitutive law reinforces the notion that fibers primarily govern the RWM’s mechanical characteristics.

Page generated in 0.0205 seconds