DOI
https://doi.org/10.25772/EV6A-KH17
Author ORCID Identifier
0000-0002-3164-3178
Defense Date
2023
Document Type
Dissertation
Degree Name
Doctor of Philosophy
Department
Physics and Applied Physics
First Advisor
Jason Reed
Second Advisor
Inho Joh
Third Advisor
Joseph Reiner
Fourth Advisor
Milos Manic
Abstract
The high-speed atomic force microscope (HS-AFM) is a device capable of collecting many full frame nanometer resolution images in real time, enabling investigation of dynamic processes and large areas not previously achievable by other means. Recent research in genomics and applied physics has shown that it is possible to diagnose certain diseases using topographic images of DNA strands immobilized on an atomically flat mica substrate, collected using an HS-AFM. This process produces diagnoses with a high degree of certainty and accuracy, but can be labor-intensive, requiring the cooperation of trained scientists of various disciplines to collect and analyze images of carefully prepared samples.
This thesis proposes a framework for streamlining this process, based on machine learning and other computer vision techniques. Exploration of this area will not only enhance the data collection and analysis process for genomics-based disease diagnosis, but more generally expand upon the growing overlap between machine learning techniques and scanning probe microscopy.
Rights
© The Author
Is Part Of
VCU University Archives
Is Part Of
VCU Theses and Dissertations
Date of Submission
12-15-2023