About
In recent years cancer has become the most common cause of death in the UK and many other developed countries. Although we now have a number of options to treat, or at least manage, the disease, the chance of success depends on how early we are able to intervene. Here we want to develop and test approaches based on epidemiology, statistics and machine learning for the early identification of people in high risk of developing different types of cancer. We will do this by identifying changes in the genome and other personal characteristics, such as life style, able to predict if someone is more likely to suffer from different types of cancer. We will then use these information to generate cancer risk calculators that can be used to assign the level of risk for each tested individual for the most common types of cancer. Finally we will test which of the factors we identified will be able to lower the risk of specified types of cancer when modified. The main aims of the work will take three years to complete, though methods development could potentially continue provided that new approaches in machine learning and statistical methodologies will emerge during this period. Our work will aim to identify those in high risk so that we can provide targeted screening for them and start intensive prevention strategies, that might not be suitable or economically viable for the general population, in order to control the identified causal factors.