Date of Graduation

12-2019

Document Type

Thesis

Degree Name

Master of Science in Biomedical Engineering (MSBME)

Degree Level

Graduate

Department

Biomedical Engineering

Advisor/Mentor

Kyle P. Quinn

Committee Member

Narasimhan Rajaram

Second Committee Member

Kartik Balachandran

Third Committee Member

Shilpa Iyer

Keywords

Biomedical imaging, Image analysis, Image quantification, Imaging techniques, Microscopy, Optics

Abstract

In biomedical optics and microscopy, the organization and morphology of organelles have been widely studied. In spite of novel imaging techniques, there is still a lack of quantitative tools to easily measure cellular characteristics from image data. Previous studies have explored multiple approaches to assess organelle organization and alignment, resulting in complicated and extensive algorithms that are both subject to multiple steps of image processing and influenced by non-cellular artifacts. In this thesis, a technique called the Modified Blanket Method (MBM) is introduced to quantify organelle organization through measurements of fractal dimension (FD) on a pixel-by-pixel basis. With the use of simulated fractal clouds, it is demonstrated that the MBM is capable of accurately and rapidly quantify FD, having a higher sensitivity to a wider range of FD values compared to previous methods. Furthermore, the MBM could differentiate mitochondrial organization of radiation-resistant A549 lung cancer cells at different time points post-radiation.

In later experiments, the MBM is combined with similar computational techniques to quantify fiber alignment and nuclear shape through measurements of directional variance (DV) and nuclear aspect ratio (NAR). The simultaneous use of these tools demonstrated that the organization and alignment of mitochondria and actin of NIH 3T3 cells treated with L-buthionine-sulfoximine (BSO) change over time, having different nuclear shapes as well. It is then concluded the this set of computational tools is capable of providing significant cellular data, which could potentially be employed to understand the cellular dynamics of multiple pathological conditions such as diabetes, Alzheimer’s, Leigh’s syndrome, and myopathy, all of which are known to be influenced by dysfunctional organelles.

Share

COinS