Output details

A multiple model approach to respiratory motion prediction for real-time IGRT

The research carried out through the Work-package co-ordinated by Haas, within the 4.5 year £7M MAESTRO EU project involving 25 partners (FP6/CE/LSHC/CT/2004/503564), was the first application of missile/aeroplane tracking to motion prediction in radiotherapy. The algorithm was evaluated on a large data set thereby demonstrating its generalisability. It introduced the 95% confidence interval which opened the debate on metric selection to evaluate prediction algorithms performance. The Kalman filter was subsequently implemented in real time within the MAESTRO couch based motion compensation system and has since been adopted and extended by other researchers. (Martin Murphy, Professor, Virginia Commonwealth University MMurphy@mcvh-vcu.edu).