Does Breathing Motion Change Weeks to Years after Treatment?
Abstract
Purpose
A CT-simulation approach has been developed which images free breathing patients with lung tumors by generating a motion model that is a function of breathing amplitude and rate (5DCT). Of the >300 patients simulated with 5DCT, 13 have been simulated more than once, two as part of a re-imaging research protocol and the others for clinical reasons, such as tumor recurrence. We compare the motion model over time.
Methods
To compare models across patients, we recast the breathing surrogate to tidal volume by using the amplitude component of the 5D motion model to deform a lung mask as a function of breathing amplitude and the volume of the mask correlated the tidal volume to the amplitude. The motion model was then recomputed using the lung volume-based surrogate and the amplitude model terms compared to both inter- and intra-patient.
Results
Of the 13 patients, three patients had more than 1000 days between their first and last treatment courses. When comparing the motion models, most changed very little despite breathing pattern changes, except for two patients with 1092 days and 461 days between their 5DCT scans. For those patients, there were marked changes of the motion model near the tumor.
Conclusion
The small model changes between scans for most patients supports the use of the model as a method for quantifying breathing motion. The change near the tumors (irradiated regions) indicates that 5DCT may be useful in the future for tumor response imaging.