Defining planning target volume in radiotherapy by using artificial neural networks

The objective of this project is to create a neural network, which generalizes doctor's knowledge and predict the planning target volume in radiotherapy from a 3-dimensional image of the detected tumor. Generally the doctor's knowledge in radiotherapy for defining the planning target volume is based on experience and some regulations. These vary significantly between doctors, institutes, diseases etc. Artificial neural networks are developed for modelling simple functions of the brain. In radiotherapy they may be of great interest because they are able to learn according to cases and regulations. In this paper the idea and the first results of predicting planning target volume by means of an artificial neural network is illustrated.