Evaluating the use of machine learning calssifiers for the identification of the determinants of stage at diagnosis of prostate cancers using registry data

This project will use a selection of Machine Learning methods to try to classify whether a given patient is diagnosed at an early or late stage of prostate cancer using a selection of registry fields (+ whatever else is identified as relevant from a gentle lit review and what can be acquired from data) as features.

How well the best model produced using each approach performs in testing will be assessed, and an outline explaining how their weightings ‘explain’ these predictions will be produced.

This a ‘methods comparison’ project in a specific domain that’s of interest to the organisation at the moment, given the uptick in diagnoses.