Abstract # 2900 Correcting For Intra-fraction Breathing Motion Based On Offline And Online Analysis Of Respiratory Patterns

Presenter: Xu, Jun

< Back to topics

> Cover Page, Authors, Institutions

> Purpose/Objective

> Materials/Methods

> Results/Discussion

> Conclusions

> View tables/images/videos

Based on patient data collected using the Varian RPM, we performed a model based offline analysis and an adaptive real-time online analysis using the Kalman filter and the linear predictive filter. In our study we found the piecewise cosine model to work better than the sinusoidal model used in previously published data. Since the piecewise cosine model is derived based on the lung motion mechanism, parameters like inhale time, exhale time and breathing cycle amplitude can be obtained through this model-based offline analysis. The linear prediction can provide more precise estimation, but is sensitive to the noise and the signal sampling frequency. These model predicted analysis of the breathing pattern allow for more precise margin definition of the target during planning and have been used in our clinic for target tracking during dynamic delivery.