Aurora Fassi1, Joël Schaerer2, Mathieu Fernandes2, Marco Riboldi3, David Sarrut2, Guido Baroni3. 1. Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Milano, Italy. Electronic address: aurora.fassi@mail.polimi.it. 2. CREATIS, CNRS UMR 5220, INSERM U1044, Université Lyon 1, INSA-Lyon, Villeurbanne, France; Department of Radiotherapy, Centre Léon Bérard, Lyon, France. 3. Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Milano, Italy; Bioengineering Unit, CNAO Foundation, Pavia, Italy.
Abstract
PURPOSE: To develop a tumor tracking method based on a surrogate-driven motion model, which provides noninvasive dynamic localization of extracranial targets for the compensation of respiration-induced intrafraction motion in high-precision radiation therapy. METHODS AND MATERIALS: The proposed approach is based on a patient-specific breathing motion model, derived a priori from 4-dimensional planning computed tomography (CT) images. Model parameters (respiratory baseline, amplitude, and phase) are retrieved and updated at each treatment fraction according to in-room radiography acquisition and optical surface imaging. The baseline parameter is adapted to the interfraction variations obtained from the daily cone beam (CB) CT scan. The respiratory amplitude and phase are extracted from an external breathing surrogate, estimated from the displacement of the patient thoracoabdominal surface, acquired with a noninvasive surface imaging device. The developed method was tested on a database of 7 lung cancer patients, including the synchronized information on internal and external respiratory motion during a CBCT scan. RESULTS: About 30 seconds of simultaneous acquisition of CBCT and optical surface images were analyzed for each patient. The tumor trajectories identified in CBCT projections were used as reference and compared with the target trajectories estimated from surface displacement with the a priori motion model. The resulting absolute differences between the reference and estimated tumor motion along the 2 image dimensions ranged between 0.7 and 2.4 mm; the measured phase shifts did not exceed 7% of the breathing cycle length. CONCLUSIONS: We investigated a tumor tracking method that integrates breathing motion information provided by the 4-dimensional planning CT with surface imaging at the time of treatment, representing an alternative approach to point-based external-internal correlation models. Although an in-room radiograph-based assessment of the reliability of the motion model is envisaged, the developed technique does not involve the estimation and continuous update of correlation parameters, thus requiring a less intense use of invasive imaging.
PURPOSE: To develop a tumor tracking method based on a surrogate-driven motion model, which provides noninvasive dynamic localization of extracranial targets for the compensation of respiration-induced intrafraction motion in high-precision radiation therapy. METHODS AND MATERIALS: The proposed approach is based on a patient-specific breathing motion model, derived a priori from 4-dimensional planning computed tomography (CT) images. Model parameters (respiratory baseline, amplitude, and phase) are retrieved and updated at each treatment fraction according to in-room radiography acquisition and optical surface imaging. The baseline parameter is adapted to the interfraction variations obtained from the daily cone beam (CB) CT scan. The respiratory amplitude and phase are extracted from an external breathing surrogate, estimated from the displacement of the patient thoracoabdominal surface, acquired with a noninvasive surface imaging device. The developed method was tested on a database of 7 lung cancerpatients, including the synchronized information on internal and external respiratory motion during a CBCT scan. RESULTS: About 30 seconds of simultaneous acquisition of CBCT and optical surface images were analyzed for each patient. The tumor trajectories identified in CBCT projections were used as reference and compared with the target trajectories estimated from surface displacement with the a priori motion model. The resulting absolute differences between the reference and estimated tumor motion along the 2 image dimensions ranged between 0.7 and 2.4 mm; the measured phase shifts did not exceed 7% of the breathing cycle length. CONCLUSIONS: We investigated a tumor tracking method that integrates breathing motion information provided by the 4-dimensional planning CT with surface imaging at the time of treatment, representing an alternative approach to point-based external-internal correlation models. Although an in-room radiograph-based assessment of the reliability of the motion model is envisaged, the developed technique does not involve the estimation and continuous update of correlation parameters, thus requiring a less intense use of invasive imaging.
Authors: S Dhou; M Hurwitz; P Mishra; W Cai; J Rottmann; R Li; C Williams; M Wagar; R Berbeco; D Ionascu; J H Lewis Journal: Phys Med Biol Date: 2015-04-23 Impact factor: 3.609
Authors: Hua-Chieh Shao; Jing Wang; Ti Bai; Jaehee Chun; Justin C Park; Steve Jiang; You Zhang Journal: Phys Med Biol Date: 2022-05-24 Impact factor: 4.174
Authors: Jens Wölfelschneider; Tobias Brandt; Sebastian Lettmaier; Rainer Fietkau; Christoph Bert Journal: Biomed Res Int Date: 2014-11-30 Impact factor: 3.411