Fast 4d Modeling for Real-Time Motion Management in Radiation Therapy
B. Gao,N. Papanikolaou,X. Xu,C. Shi
DOI: https://doi.org/10.1016/j.ijrobp.2010.07.1689
2010-01-01
Abstract:Purpose/Objective(s)Respiratory motion management is a critical task in today's radiation therapy. Monitoring the three-dimensional deformable motion of structures in breathing presents a major clinical challenge. As an alternative, this study proposes a fast modeling technique which can be used to predict the real-time motion of tumor and critical organs of a patient during free breathing.Materials/MethodsA dynamic geometry model was reconstructed based on the 4D-CT scan of a patient and verified against the 4D-CT. Applying novel surface representative techniques and mathematical prediction filter, the model is capable of predicting in real-time the complex motion of tumor and critical organs under different respiratory conditions, including regular breathing with various amplitude and frequency and irregular free-breathing. The dynamic model was applied in 4D treatment planning and delivery study. For treatment planning, models were re-converted to multi-phase CT images. Monte Carlo method was used to study the dosimetric effect of motion under regular and irregular breathing. For 4D treatment delivery, a method was proposed to extract the motion of tumor from dynamic model and incorporate it into dynamic multi-leaf collimator (dMLC) sequence.ResultsThe dynamic model agreed with 4D-CT of the patient: The percent differences of structure volumes between the dynamic model and the original contour were within 5%. Center of mass (COM) of all structures agreed within 1 mm in x, y, z directions. Percent volume overlap (PVO) between the model and original contours was larger than 95% for all structures. Preliminary study also proved the potential of using dynamic model in 4D radiation therapy planning and delivery.ConclusionsA dynamic model closely represents the geometry and motion of structures for a patient. Although based on a single 4D-CT set, the model can be extended to various respiratory conditions. With real-time motion prediction capability, the dynamic model has shown potential to be used in guidance of radiation therapy treatment planning and delivery. Purpose/Objective(s)Respiratory motion management is a critical task in today's radiation therapy. Monitoring the three-dimensional deformable motion of structures in breathing presents a major clinical challenge. As an alternative, this study proposes a fast modeling technique which can be used to predict the real-time motion of tumor and critical organs of a patient during free breathing. Respiratory motion management is a critical task in today's radiation therapy. Monitoring the three-dimensional deformable motion of structures in breathing presents a major clinical challenge. As an alternative, this study proposes a fast modeling technique which can be used to predict the real-time motion of tumor and critical organs of a patient during free breathing. Materials/MethodsA dynamic geometry model was reconstructed based on the 4D-CT scan of a patient and verified against the 4D-CT. Applying novel surface representative techniques and mathematical prediction filter, the model is capable of predicting in real-time the complex motion of tumor and critical organs under different respiratory conditions, including regular breathing with various amplitude and frequency and irregular free-breathing. The dynamic model was applied in 4D treatment planning and delivery study. For treatment planning, models were re-converted to multi-phase CT images. Monte Carlo method was used to study the dosimetric effect of motion under regular and irregular breathing. For 4D treatment delivery, a method was proposed to extract the motion of tumor from dynamic model and incorporate it into dynamic multi-leaf collimator (dMLC) sequence. A dynamic geometry model was reconstructed based on the 4D-CT scan of a patient and verified against the 4D-CT. Applying novel surface representative techniques and mathematical prediction filter, the model is capable of predicting in real-time the complex motion of tumor and critical organs under different respiratory conditions, including regular breathing with various amplitude and frequency and irregular free-breathing. The dynamic model was applied in 4D treatment planning and delivery study. For treatment planning, models were re-converted to multi-phase CT images. Monte Carlo method was used to study the dosimetric effect of motion under regular and irregular breathing. For 4D treatment delivery, a method was proposed to extract the motion of tumor from dynamic model and incorporate it into dynamic multi-leaf collimator (dMLC) sequence. ResultsThe dynamic model agreed with 4D-CT of the patient: The percent differences of structure volumes between the dynamic model and the original contour were within 5%. Center of mass (COM) of all structures agreed within 1 mm in x, y, z directions. Percent volume overlap (PVO) between the model and original contours was larger than 95% for all structures. Preliminary study also proved the potential of using dynamic model in 4D radiation therapy planning and delivery. The dynamic model agreed with 4D-CT of the patient: The percent differences of structure volumes between the dynamic model and the original contour were within 5%. Center of mass (COM) of all structures agreed within 1 mm in x, y, z directions. Percent volume overlap (PVO) between the model and original contours was larger than 95% for all structures. Preliminary study also proved the potential of using dynamic model in 4D radiation therapy planning and delivery. ConclusionsA dynamic model closely represents the geometry and motion of structures for a patient. Although based on a single 4D-CT set, the model can be extended to various respiratory conditions. With real-time motion prediction capability, the dynamic model has shown potential to be used in guidance of radiation therapy treatment planning and delivery. A dynamic model closely represents the geometry and motion of structures for a patient. Although based on a single 4D-CT set, the model can be extended to various respiratory conditions. With real-time motion prediction capability, the dynamic model has shown potential to be used in guidance of radiation therapy treatment planning and delivery.