Decision trees
Decision trees are the simplest models. The essence of this technique is contained in its name ? nodes and branches represent possible pathways of the patient?s management.
A decision tree consists of three types of nodes:
- decision nodes - at which further action depends on a decision (e.g. a physician's decision on the second-line treatment to be used following the failure of a first-line treatment or a patient's decision to accept or refuse surgical treatment),
- chance nodes - further action depends on an event occurring with a specific probability (e.g. whether the patient will survive or die as a result of the therapy applied),
- end nodes - terminating the pathway (e.g. cure or death).
Flexibility of the decision tree makes it possible to add or eliminate branches in order to accurately represent the reality (i.e. to make the modelled problem more or less complicated). A decision tree may be a base for further simulation, e.g. in Markov models.