I enjoy making things. Here are a selection of projects that I have worked on over the years.
I was part of the development of this important Personalized Sarcoma Care (PERSARC) prediction tool, with the aim of improving care of patients with high-grade soft tissue sarcoma. Multiple statistical models have been developed and updated as the project grew larger and more international Data could be collected. The models are available for clinical practice through an app for androis and ios.
I contributed to several projects that applied multi-state models to clinical data. One key project focused on identifying risk factors for patients with Ewing sarcoma. A multi-state model, like the one shown on the left, can capture multiple health events a patient may experience. It allows for the modeling of risk factors associated with each transition between different “states” in a patient’s health journey. Additionally, such models enable predictions for individual patients, helping inform clinical decision-making.
A dynamic prediction model for survival of patients with soft tissue sarcoma. After surgery patients can experience disease related events, such as local recurrence (LR) that change their predicted probability of survival. Even without experiencing such events, the fact that a patient is alive at e.g. 1 year past surgery, can have an effect on his survival probability. A dynamic prediction model is able to model the patient journey as they unfold in clinical practice. On the left, Panel C of Figure 2 shows how these probabilities change over time for a particular patient (Figure 2C. 5-year probability of death estimates for patients with different characteristics and at different states of disease progression (Rueten-Budde et al., 2018). Reprinted from Surgical Oncology, 27(4), 695-701. (https://doi.org/10.1016/j.suronc.2018.09.003)).
