"4D" imaging for motion - WP5: Simulation et modélisation d'images Accéder directement au contenu
Communication Dans Un Congrès Année : 2014

"4D" imaging for motion

Résumé

In the treatment of moving targets with proton beams, time resolved image guidance is required for the effective compensation of geometrical uncertainties due to motion. Associated density and radiological variations in path length need to be accurately modeled at the treatment planning stage, and adequately captured at treatment delivery. The impact of motion/density changes on the dose distribution is larger than for conventional X-ray treatments. The ideal system for 4D image guidance in proton therapy should provide complete anatomical and radiological path length information during treatment, with the highest spatial and temporal resolution. In this scenario, the availability of reliable and representative 4D imaging datasets for treatment planning is a key issue in order to provide robust plans. Specifically, artifact-free 4D CT volumes are required to model density changes in the beam eye view, and to depict motion throughout the breathing cycle. Accurate motion quantification is also needed to adequately account for motion in 4D treatment planning. The validation of quantified motion models, as well as the development of automated procedures for quantification and/or validation, is still a challenge for clinical application. Dedicated imaging protocols in MRI or US may also endow breathing motion description under irregular respiration, thus allowing the test of plan robustness against breathing variability. Besides technological restrictions and current availability of in-room imaging solutions, effective time resolved image guidance in treatment delivery is intrinsically limited by the need to trade-off between accuracy and imaging dose. Similar considerations motivated the simultaneous use of external surrogates and external/internal correlation models in conventional X-ray radiotherapy, as a way to achieve viable "4D" imaging protocols during treatment delivery. A specific extension of model based "4D" imaging to proton therapy requires the integration of radiological path length variations within the estimated/measured variables. Technological solutions to measure path length information in-room are currently being developed, thus potentially providing adequate means to verify model-based estimates. Four-dimensional imaging is also expected to play a role in proton therapy treatment verification, as the distribution of the delivered dose can be, to some extent, measured in vivo, either during or immediately after treatment delivery (e.g. with PET or prompt-gamma). In this case, specific challenges in 4D imaging are represented by the extremely reduced statistics of the available data for image reconstruction, leading to extremely noisy 4D datasets, where motion quantification is demanding. In this contribution, "4D"imaging modalities for treatment planning and delivery in proton therapy will be overviewed, with a specific focus on motion quantification. The integration between imaging data and appropriate modeling approaches will be discussed, aiming at the assessment of the intrinsic accuracy of current approaches to "4D" imaging in particle therapy.

Domaines

Imagerie
Fichier non déposé

Dates et versions

hal-01053174 , version 1 (29-07-2014)

Identifiants

  • HAL Id : hal-01053174 , version 1

Citer

M. Riboldi, C. Bert, K. Parodi, D. Sarrut, G. Baroni. "4D" imaging for motion. ESTRO 33, Apr 2014, Vienne, Austria. pp.SP-0222. ⟨hal-01053174⟩
132 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More