Project title: Developing a risk score for pancreatic cancer
diagnosis using machine learning techniques applied to
linked routine data: a pilot study
Project aims: Dr Woods’s project addresses the challenge of diagnosing pancreatic cancer earlier. She will apply machine learning techniques to an historic, anonymised database of thousands of GP records to examine whether people who later developed pancreatic cancer shared similar early warning signs detectable before diagnosis. If successful, this could provide means of identifying a population of patients whom it would be cost-effective to screen. This would increase the number of cancers diagnosed at a treatable stage.