Research theme 11
Digital Twins & In-Silico Oncology
Longitudinal cancer models, integrating evolving patient context, and the distinction between computational modelling and established clinical evidence.
Stated area of interest Named by KPCIRC as an area of scientific interest. Not a claim of an active funded project.
A digital twin, in this context, is an evolving computational model of a person’s cancer journey that brings molecular, clinical, imaging and treatment information into one longitudinal view. This theme examines how such models can organise information without being mistaken for clinical evidence.
Questions of interest
- Which data are needed to keep a longitudinal model current and accurate?
- How should model outputs be labelled so they are not read as predictions of benefit?
- How can model assumptions be made visible to clinicians?
Approaches
- Longitudinal data models with provenance
- Explicit research-use labelling
- Separating observed data from modelled estimates
Connection to clinical services
Connected to OnKommon TWIN, which is described as Research Use Only where labelled.
Limitations
- Computational models are not established clinical evidence.
- A modelled scenario is not a prediction of an individual’s outcome.
Projects and outputs
No verified projects or publications are listed for this theme yet.