Recurrent neural network for tumor movement prediction

A system for predicting respiratory motion in real time, the system comprising:

• a system controller circuitry;
• a training module;
• and
• a recurrent neural network circuitry,
• wherein the training module is configured to train the recurrent neural network circuitry in real time, the training is performed over a training time interval, the training is based, at least in part, on a training data set, the training data set comprises a training number of measured position samples, and the measured position samples are related to respiratory motion of a patient target region;
• and
• the trained recurrent neural network circuitry is configured to predict a future position of the patient target region at a future point in time based, at least in part, on a prediction data set, the future point in time is a look ahead time interval in the future relative to a prediction time interval, and the prediction data set comprises a prediction number of measured position samples.