Longitudinal Health Transformer for Cancer Pathways Modelling

Gerrard, Leah B.

Longitudinal Health Transformer for Cancer Pathways Modelling

Gerrard, Leah B.

Permalink

Publication Type:: Thesis
Issue Date:: 2025

Open Access

Copyright Clearance Process

Recently Added
In Progress
Open Access

This item is open access.

Adobe PDF

Download thesisAdobe PDF (1.98 MB)

View statistics

Full metadata record

Field	Value	Language
dc.contributor.author	Gerrard, Leah B.
dc.date.accessioned	2026-06-23T04:28:13Z
dc.date.available	2026-06-23T04:28:13Z
dc.date.issued	2025
dc.identifier.uri	http://hdl.handle.net/10453/195448
dc.description	University of Technology Sydney. Faculty of Engineering and Information Technology.	en_US.UTF-8
dc.description.abstract	The success of artificial intelligence in modelling complex data, from natural language to visual inputs, has inspired its application to healthcare. In cancer care, the cancer pathway consists of a sequence of longitudinal health data that combines development of the disease with interactions between patients and various healthcare providers. Modelling both the disease progression and human interactions within cancer pathways remains a significant challenge. As a result, there is a pressing need for approaches that can better capture and understand complex cancer data and pathways for unified solutions. This includes considering the unique characteristics of cancer pathways being multi-outcome, captured within various data sources, and often suffering from limited data and labels. To address these issues, this research focuses on developing deep learning-based methods for cancer pathways modelling, utilising transformer-based models with longitudinal health data. A series of models are proposed that leverage strategies including multi-task learning; longitudinal patient modelling; and transfer learning; that are tailored to the cancer context. Experimental testing of these models demonstrates their ability to improve predictions for cancer patients and provide more effective, flexible, and data-efficient approaches. This work illustrates the value of the Transformer in capturing intricate relationships between cancer patients and the healthcare system, offering a promising foundation for advancing the modelling of cancer pathways and improving the care and outcomes for cancer patients.	en_US.UTF-8
dc.format	Thesis (PhD)
dc.language.iso	en_US	en_US.UTF-8
dc.relation	https://opus.lib.uts.edu.au/bitstream/10453/195448/1/thesis.pdf
dc.rights	info:eu-repo/semantics/openAccess
dc.rights	The author owns the copyright in this thesis including all reproduction and reuse rights for the work. The work may not be altered without the permission of the copyright owner. Attribution is essential when quoting or paraphrasing from this thesis.
dc.rights	© 2025 Leah B. Gerrard
dc.rights	au.edu.uts.lib/cph
dc.title	Longitudinal Health Transformer for Cancer Pathways Modelling	en_US.UTF-8
dc.type	Thesis
utslib.copyright.status	open_access	*

Abstract:

The success of artificial intelligence in modelling complex data, from natural language to visual inputs, has inspired its application to healthcare. In cancer care, the cancer pathway consists of a sequence of longitudinal health data that combines development of the disease with interactions between patients and various healthcare providers. Modelling both the disease progression and human interactions within cancer pathways remains a significant challenge. As a result, there is a pressing need for approaches that can better capture and understand complex cancer data and pathways for unified solutions. This includes considering the unique characteristics of cancer pathways being multi-outcome, captured within various data sources, and often suffering from limited data and labels. To address these issues, this research focuses on developing deep learning-based methods for cancer pathways modelling, utilising transformer-based models with longitudinal health data. A series of models are proposed that leverage strategies including multi-task learning; longitudinal patient modelling; and transfer learning; that are tailored to the cancer context. Experimental testing of these models demonstrates their ability to improve predictions for cancer patients and provide more effective, flexible, and data-efficient approaches. This work illustrates the value of the Transformer in capturing intricate relationships between cancer patients and the healthcare system, offering a promising foundation for advancing the modelling of cancer pathways and improving the care and outcomes for cancer patients.

Please use this identifier to cite or link to this item:

http://hdl.handle.net/10453/195448