Georgia Institute of Technology
Spatial AI analysis of single-cell image-based omics data
Abstract
dc:description.abstractA diverse array of spatial learning methods tailored for analyzing spatially resolved singlecell multiplex omics data are introduced. Initially, SpatialViz pipelines are introduced, which serve as foundational tools for analyzing disease and health dynamics at the single-cell level within their spatial contexts. By expanding the repertoire of protein targets, SpatialVizScore and SpatialVizPheno enable the uncovering of multiple interaction types across patient tissues. Subsequently, the integration of metabolomic and protein data is presented as crucial for understanding metabolomic changes at the single-cell level, offering insights into cellular metabolism and its implications for disease pathogenesis and treatment response. As the adoption of spatial omics technologies increases across diseases, integrating multi-omics data becomes imperative for deciphering underlying biological processes. Lastly, graph-based deep learning pipelines are described for analyzing single-cell and subcellular omics data, offering a powerful framework for capturing complex relationships within spatially resolved datasets. These pipelines demonstrate the potential to extract disease biomarkers and predict treatment responses accurately and help with cases of missing or lower-quality multiplex stains.
Degree
thesis:*- Level thesis:degree_level
- Doctoral
- Department dc:contributor.department
- Electrical and Computer Engineering
- Grantor dc:publisher
- Georgia Institute of Technology
- Year dc:date.issued
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Hu, Thomas Kaiwen
- Advisors dc:contributor.advisor
-
- Coskun, Ahmet F.
- Inan, Omer T.
- Committee members dc:contributor.committeemember
-
- Dyer, Eva L.
- Qiu, Peng
- Farris, Alton B.
- Mei, Yajun
Subjects
dc:subject × 3Rights
- Language dc:language.iso
- en_US
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- https://hdl.handle.net/1853/77693