Analysis of Cancer Margins in Mass Spectrometry Images

Computational analysis of metabolite activity surrounding a tumor border could enhance intraoperative tumor resection. One method to collect this data is mass spectrometry imaging, an analytical technique that can detect the patterns of metabolites within a sample and retain the spatial information. The patterns of metabolites can provide further insight into the functioning of cells. Our data included mass spectrometry images of excised tissue samples from patients who underwent the removal of skin and breast cancer. By applying computational analysis to a pathologist-defined tumor boundary, we compared patterns of metabolites to the physical annotations. We proposed a linear model that reconstructed mass spectrometry signals from the tumor boundary using only tumor and non-tumor signals. By creating a linear combination of both tumor and non-tumor signals, we could determine the type of metabolite activity surrounding the tumor boundary. We hypothesized a linear relationship between metabolite signals along the tumor border and between tumor and non-tumor regions. Our analysis strongly supported our hypothesis. The results suggested that the tumor stroma was a biologically active zone that extended past the pathologically annotated boundary in every sample.

Department dc:contributor.department

Computing

Year dc:date.issued

2024

Author dc:creator

• Katherine Williams

Advisor dc:contributor.supervisor

• Ellis, Randy

• Mass Spectrometry Imaging
• Breast Cancer
• Skin Cancer
• Biomedical Computing
• Tumor Stroma
• Machine Learning
• Computing

Statement dc:rights

• Attribution-NonCommercial 4.0 International

Licence dc:rights.uri

• http://creativecommons.org/licenses/by-nc/4.0/

Language dc:language.iso

eng

Handle dc:identifier.uri

https://hdl.handle.net/1974/33294

OAI identifier oai:identifier

oai:queensu.scholaris.ca:1974/33294