Work

Quantitative Prognostic Stromal Assessment of Colorectal Cancer using Polarized Light Microscopy

Year: 2023
Type: dissertation
Study design: Observational Study
Abstract: Colorectal cancer remains the second-leading cause of cancer-related deaths in North America. In recent years, patients have observed an overall improvement in survival, however, a significant portion still experience unfavourable outcomes as consequence of high recurrence rates. An effort to mitigate this issue is by assessing the prognostic value contained within collagen surrounding the tumour, where it has been shown to correlate well with prognostic outcomes. However, its clinical uptake is withheld by its inherent subjective nature and lack of standardized approach. This thesis employs polarized light microscopy to address some of these challenges. It expands upon previously published work by introducing texture analysis and unsupervised machine learning. These additions and associated findings are presented in the form of a published study analyzing correlations between this revised polarimetric framework with known patient survival outcomes. Furthermore, this thesis contains preliminary work which was instrumental in the lead-up to the primary study along with additional analyses and discussion looking to improve upon the published findings. Overall, the work presented herein serves as a promising beginning for alternative methods of collagen assessment and for the use of more complex machine learning methods for its prognostic evaluation.
Author: Lad, Jigar
Language: English
Open Access status: green
OpenAlex ID: W7132985801
Cites: 59
Cited by:
Related to: 10