Date of Award
Fourier transform, Frequency analysis, Isotropy, Trabecular bone
Medical data is hard to obtain due to privacy laws making research difficult. Many databases of medical data have been compiled over the years and are available to the scientific community. These databases are not comprehensive and lack many clinical conditions. Certain type of medical conditions are rare, making them harder to obtain, or are not present at all in the aforementioned databases. Due to the sparsity or complete lack of data regarding certain conditions, research has stifled. Recent developments in machine learning and generative neural networks have made it possible to generate realistic data that can overcome the lack of data in the medical field. For example, the study of osteoporosis, poor bone density, or osteoarthritis, over abundance of bone material, has been studied extensively. Despite the commonality of these conditions, getting data is still difficult. This data is required to develop treatments but it is a problem that continues to plague an ever aging population. The ability to generate bone data with specific micro-structural parameters, enabling the simulation of real world data, for a desired medical condition would remove the high barrier that currently exists in the field.
This thesis explores fundamental properties of bone microstructure that can help describe said structure. This would provide insight into the bone spongiosa, the interior portion of bones, which contributes to bone stability. It explores the bone microstructure in the frequency space through the Fourier transform in order to explore useful properties, like isomorphism.
Copyright is held by the author. User is responsible for all copyright compliance.
Parada San Martin, Daniel, "Frequency Analysis of Trabecular Bone Structure" (2022). Electronic Theses and Dissertations. 2143.
Received from ProQuest
Daniel Parada San Martin