Deciphering and modelling the action of immune cells using highly multiplexed imaging and deep learning techniques

Cells of the immune system are capable of responding to foreign antigen, promoting host defense while limiting damage to host tissues, through an act known as selftolerance. T cells, their activation and their effector roles are of particular interest due to their prominent roles in antigen discrimination and subsequent cell-mediated immunity. However, there are diverse effector T cell types interacting to regulate the immune response. Understanding the mechanisms by which intercellular interactions exert precise control over the immune system is a crucial step in elucidating the manner in which the immune system behaves during infection, health, or chronic disease. Multiplexed imaging is a beneficial tool that is used to visualize distinct cell types and functional states directly in tissues. This technology is particularly important for understanding how cells organize spatially to enforce this boundary between hostprotective responses and autoimmunity. Therefore, it is valuable to image interacting cells in highly-multiplexed images. In order to do this, it has become increasingly important to increase the number of biomarkers that one can record in a single tissue section at a time. Here, I summarize our efforts to employ imaging and deep learning tools to analyze the structure of the immune system, ending with a critical insight regarding our cell segmentation models alongside an experimental workflow and pipeline that will allow even more to be revealed about the mechanisms of control that exist within the immune system. Current methods for acquiring highly multiplexed images are somewhat time-consuming and labour-intensive while computational methods for analyzing these images and identifying relevant spatial patterns are lacking. We seek to improve and simplify our current multiplexing capabilities by eventually coupling fluorescence lifetime with fluorescence intensity measurements—two distinct imaging modalities. Moreover, we aim to develop new computational pipelines to aid in downstream image analysis and identify new spatial motifs that control immune response in tissues.

Name thesis:degree_name

Master

Department dc:contributor.department

Massachusetts Institute of Technology. Department of Electrical Engineering and Computer Science

Grantor dc:publisher

Massachusetts Institute of Technology

Year dc:date.issued

2023

Author dc:creator

• Reid, Clinton

Advisor dc:contributor.advisor

• Wong, Harikesh

Statement dc:rights

• In Copyright - Educational Use Permitted
• Copyright retained by author(s)

Licence dc:rights.uri

• https://rightsstatements.org/page/InC-EDU/1.0/

Handle dc:identifier.uri

https://hdl.handle.net/1721.1/151517

OAI identifier oai:identifier

oai:dspace.mit.edu:1721.1/151517