Back to search

Cornell University

TWO TOOLS FOR THREE CHARACTERIZATIONS OF CELLS: MACHINE LEARNING FOR AUTOMATED CANCER CELL IDENTIFICATION AND ELECTROROTATION FOR CANCER CELL AND ALGAE CELL ELECTRICAL PROPERTY MEASUREMENT

Abstract

dc:description.abstract

Advances in rare cell capture technology have made possible the interrogation of circulating tumor cells (CTCs) captured from patient blood. However, manually locating captured cells in the device bottlenecks data processing by being tedious (hours per sample) and compromises the results by being inconsistent and prone to user bias. In the first aim of my thesis, I employed machine learning algorithms to locate and classify thousands of possible cells in a few minutes rather than a few hours, representing an order of magnitude increase in processing speed performance. Optimal algorithm selection depends on the peculiarities of the individual dataset, indicating the need of a careful comparison and optimization of algorithms for individual image classification tasks. The capture of circulating tumor cells via immuno-affinity may be compromised by reduced antigen expression associated with acquired resistance to chemotherapy, deprivation of growth factors when entering circulation, and the epithelial-to-mesenchymal transition. Dielectrophoresis (DEP), however, could enhance the capture of these rare cells by attracting cells to the antibody-coated surface. In order to reliably use dielectrophoresis, cancer cell crossover frequencies must remain lower than those of white blood cells. For my second aim, I used automated electrorotation to measure the cytoplasmic permittivity, cytoplasmic conductivity, and specific membrane capacitance of pancreatic cancer cells under three treatments: 1) acquired resistance to gemcitabine, 2) serum starvation, and 3) induced epithelial-to-mesenchymal transition. I found that the median computed crossover frequency for cancer cells under all treatments remains significantly below that of blood cells, indicating that DEP is a promising technique for enhancing capture. Algae are a promising feedstock for biofuels, and there is a critical need for a rapid, inexpensive, and label-free measurement of lipid accumulation in algae cells. Measuring the electrical properties of algae has shown promise in the literature for monitoring lipid accumulation because it correlates with a decrease in effective cytoplasmic conductivity. Previous models, however, have assumed a constant cytoplasmic permittivity through the lipid accumulation process, and that assumption must be validated. For my third aim, I used automated electrorotation to measure properties of Chlamydomonas reinhardtii cells undergoing lipid accumulation.

Degree

thesis:*
Name thesis:degree_name
Ph. D., Mechanical Engineering
Level thesis:degree_level
Doctor of Philosophy
Discipline thesis:degree_discipline
Mechanical Engineering
Grantor
Cornell University
Year dc:date.issued
2017

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Lannin, Timothy Burke
Chair dc:contributor.chair
  • Kirby, Brian
Committee members dc:contributor.committeemember
  • Singh, Ankur
  • Feigenson, Gerald W

Subjects

dc:subject × 7

Rights

Language dc:language.iso
en_US

Identifiers

dc:identifier.*
Dc Identifier Bibid
9906082
Dc Identifier Other
ProQuest Submission ID: 10026
ProQuest Publication ID: 10174000

Chain of custody

source
Harvested from
Cornell University
Base URL
ecommons.cornell.edu/server/oai/request
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Lannin, Timothy Burke. TWO TOOLS FOR THREE CHARACTERIZATIONS OF CELLS: MACHINE LEARNING FOR AUTOMATED CANCER CELL IDENTIFICATION AND ELECTROROTATION FOR CANCER CELL AND ALGAE CELL ELECTRICAL PROPERTY MEASUREMENT. Doctor of Philosophy thesis, Cornell University, 2017. https://hdl.handle.net/1813/47835