Digital Twins & Predictive Oncology
Computational replicas of a tumour that forecast behaviour and likely response before decisions are taken in clinic.
What this centre does
Builds computational replicas of a patient's tumour that can forecast how a cancer is likely to behave and how it may respond to different treatment strategies, before those decisions are made in the clinic.
Molecular data via OnKommon
Methods
Mechanistic and statistical modelling of tumour growth and response; multi-modal machine learning; survival and time-to-event modelling; simulation of treatment scenarios; uncertainty quantification.
Data inputs
Longitudinal molecular profiles and serial ctDNA (via OnKommon); staging and response imaging; treatment and outcome data from the clinic; published cohort data.
Outputs
Validated predictive models; risk-stratification tools for research use; publications.
Funding route
AI-in-health and computational biology grant calls; industry collaborations on response prediction.
What has to be true first
Sufficient longitudinal cases with serial molecular data — realistically a 24-month accumulation before meaningful modelling.
Themes this centre carries
Work with this centre
The research centres collaborate with funders, sponsors, hospitals and academic groups. Tell us what you are working on and the enquiry is routed to the centre lead.
Contact the research office