Development of digital holographic vibrometry methods for determination of mechanical properties of bioengineered tissues


Downloadable Content

open in viewer

Modern development of bioengineered tissues for therapeutic applications calls for the advancement of methodologies to increase the efficiency of the steps used in large-scale manufacturing. One of the steps in the manufacturing processes that requires further development is quality control where determination of physical properties is currently slow, labor intensive, destructive, and requiring improved precision. This Thesis describes the creation, validation, and preliminary application of a digital holographic vibrometry (DHV) sensor to measure the mechanical properties of tissues. This sensor uses acoustic energy from a sound source to induce vibrations in tissue samples together with high-resolution digital holography methods. Full-field-of-view, real-time modes of vibration of samples are obtained with micrometer and nanometer spatial and displacement measurement resolutions, respectively. The mode shapes and the frequencies at which natural modes of vibration occur are correlated to the mechanical properties of tissue samples. The tissue used in this Thesis is Apligraf biofabricated skin, manufactured by Organogenesis Inc. Using Apligraf as an initial application, experimental modal analyses are conducted together with Finite Element (FE) simulations to train a machine learning algorithm. This Thesis demonstrates the capabilities of such approach to recover mechanical properties of intact packaged tissues noninvasively. Validation is supplemented by application of optical coherence tomography (OCT) and nanoindentation measurements. Further, this Thesis discusses the application of the sensor to a string model. This serves as a proof of concept of the applicability of the sensor to noninvasively investigate bioengineered ligaments as well as potentially other bioengineered tissues. This Thesis finds the presented methodologies to successfully noninvasively measure vibrations in the two tissue models. However, complexities in the packaging make it difficult to identify mechanical properties from these vibrations using standard analytical methods. As such a machine learning approach is under present development to interpret the collected data and relate vibration shape and frequency to the mechanical properties of the tissue. Future developments of the methodologies can enable the realization of instrumentation and algorithms for quality control of bioengineered tissues in large-scale manufacturing settings.

  • etd-64356
Defense date
  • 2022
Date created
  • 2022-04-27
Resource type
Rights statement
Last modified
  • 2022-09-09


In Collection:



Permanent link to this page: