ENDORISk is a personalized model which can predict the risk of lymph node metastases and survival in patients with endometrial carcinoma. Several diagnostic variables can be used in the model, such as pathology, imaging and tumor markers.

 

ENDORISK is a Bayesian network. This is a probabilistic model, which means that the model calculates the chance of an outcome for each separate variable. The outcomes in ENDORISK are lymph node metastases and survival.

​

The model also calculates correlations between variables. This enables the possibility to model and analyse dependent and independent relations between variables. Variables that do not have a high individual predictive quality can have a higher collective predictive quality together with other variables. Another advantage of a Bayesian network is that it is not necessary to input all variables to calculate an outcome.