- Email: [email protected]
CT Treatment Verification after Proton Therapy in Abdominopelvic Tumor Sites
I. J. Radiation Oncology d Biology d Physics
S642
3051
Volume 72, Number 1, Supplement, 2008
PET/CT Treatment Verification after Proton Therapy in Abdominopelvic Tumor Sites 1
H. A. Shih , A. Knopf1, K. Parodi2, H. Paganetti1, T. Bortfeld1 1
Massachusetts General Hospital, Boston, MA, 2Heidelberg Ion Therapy Center, Heidelberg, Germany
Purpose/Objective(s): A promising method for in vivo, non-invasive monitoring of radiation treatments with proton beams is positron emission tomography (PET). Previous studies at Massachusetts General Hospital showed the potential for mm-accurate range verification for tumors in the cranial base and spine. In this study we investigate whether offline PET/CT scans are also suitable for accurate range verification in abdominopelvic tumor sites which are challenged by patient motion and tissue heterogeneity in both composition and biological washout characteristics. Materials/Methods: PET data were acquired for 30min in list mode starting within 20min after irradiation using a commercial PET/CT scanner. The measured PET distributions were compared to simulations of the PET signal based on Geant4 and FLUKA Monte Carlo (MC) codes. To investigate the effects of patient motion, the lateral agreement between measured and simulated activity distributions for study cases were compared to the lateral conformity in head and neck tumor cases. Hounsfield Unit (HU) domains for different organs in the abdomen and pelvis were determined to investigate whether the HU mapping is sufficiently correlative to assign organ specific elemental compositions and washout characteristics in the MC simulations. Range comparisons between measurement and simulation at specific positions in the fall-off region of the activity (point wise verification) as well as a range analysis taking the entire fall-off region into account (shift verification) were performed. Analyses were based upon three pelvic and one spinal tumor cases. Results: Lateral blurring in abdominopelvic tumor sites ranged up to 25mm compared to prior documented conformity for head and neck tumor sites of up to 5mm. Lateral blurring is dominant in the anterior-posterior direction reflecting the breathing motion during the 30min of PET scan. The HU analysis showed overlapping HU domains, for example in bone marrow and soft tissue. This prevents an organ specific assignment based upon density and thus does not provide an accurate simulation of the total yield and spatial distribution of b+-activity. Mean range deviation between measurement and simulation in the prostate cases at the 20% and 50% positions in the last activity fall-off was 3.1mm and 5.4mm, respectively, and was 4.5mm by shift verification. For opposed treatment beams, point wise as well as shift range verification were found to be impractical. Conclusions: In abdominopelvic tumor sites, mm-accurate offline PET/CT range verification is not currently feasible primarily due to patient motion and the position of the distal beam edge in soft tissue. Author Disclosure: H.A. Shih, None; A. Knopf, None; K. Parodi, None; H. Paganetti, None; T. Bortfeld, None.
3052
Low-dose X-ray Fluoroscopy for Image Guided Radiation Therapy (IGRT)
X. Zhang, J. Wang, L. Zhu, L. Xing Stanford University School of Medicine, Stanford, CA Purpose/Objective(s): X-ray fluoroscopic imaging plays an important role in IGRT. There are growing interests in using X-ray fluoroscopy for the management of organ motion and gating in radiotherapy. However, long-time exposure of X-rays may lead to adverse health effect to the patients. In this work, we aim to improve the quality of low-dose X-ray fluoroscopic images using statistics-based restoration algorithm so that the patient fluoroscopy can be performed with much reduced radiation dose. Method/Materials: The noise in the low-dose fluoroscopy was suppressed by temporal plus spatial filtering. The correlation among neighboring frames was considered by the Karhunen-Loe`ve (KL) transform. After the KL transform, the selected neighboring frames of fluoroscopy was decomposed to un-correlated and ordered principal components; and each KL component was associated with an Eigen value. For each KL component, a penalized weighted least-squares (PWLS) objective function was constructed to restore the ideal image. The PWLS criterion reflects that the measurement with smaller signal-to-noise ratio (SNR) will contribute less to the objective function. The variance of each measurement was chosen as the weight since the measured signal can be modeled as Poisson noise and the measurement of larger variance has larger SNR for Poisson noise. The penalty was chosen as quadratic and the penalty parameter in each KL component was inverse proportional to its corresponding Eigen value. Smaller KL Eigen value is associated with the KL component of lower SNR, and a larger penalty parameter should be used for such KL component. The low-dose fluoroscopic images were acquired by a Varian Acuity simulator and a quality assurance phantom was used to evaluate the presented algorithm. Results: Low-dose fluoroscopic images were acquired with X-ray tube current of 10 mA and duration of X-ray pulse 2 ms. In the image restored by the proposed KL domain PWLS algorithm, noise is greatly suppressed while fine structures are well preserved. Comparison studies with traditional techniques, such as the mean and median filters, show that the proposed algorithm is advantageous in terms of resolution preservation. Conclusions: The proposed noise reduction algorithm can significantly improve the quality of low-dose X-ray fluoroscopic image and allows for dose reduction in X-ray fluoroscopy. Author Disclosure: X. Zhang, None; J. Wang, None; L. Zhu, None; L. Xing, None.
3053
Clinical use of Electromagnetic Guidance for Lung and Spine Radiation Therapy
T. R. Willoughby, A. P. Shah, A. R. Forbes, R. R. Manon, P. A. Kupelian, S. L. Meeks MD Anderson Cancer Center, Orlando, Orlando, FL Purpose/Objective(s): To describe the first clinical use of the Calypso 4D Localization System in regions outside of the prostate gland, specifically in spine and lung radiotherapy. Materials/Methods: The Calypso System (Calypso Medical, Seattle, WA) is approved for routine guidance for external beam radiotherapy of localized prostate cancers. However, other anatomic sites could benefit from similar guidance of high dose external
S642
3051
Volume 72, Number 1, Supplement, 2008
PET/CT Treatment Verification after Proton Therapy in Abdominopelvic Tumor Sites 1
H. A. Shih , A. Knopf1, K. Parodi2, H. Paganetti1, T. Bortfeld1 1
Massachusetts General Hospital, Boston, MA, 2Heidelberg Ion Therapy Center, Heidelberg, Germany
Purpose/Objective(s): A promising method for in vivo, non-invasive monitoring of radiation treatments with proton beams is positron emission tomography (PET). Previous studies at Massachusetts General Hospital showed the potential for mm-accurate range verification for tumors in the cranial base and spine. In this study we investigate whether offline PET/CT scans are also suitable for accurate range verification in abdominopelvic tumor sites which are challenged by patient motion and tissue heterogeneity in both composition and biological washout characteristics. Materials/Methods: PET data were acquired for 30min in list mode starting within 20min after irradiation using a commercial PET/CT scanner. The measured PET distributions were compared to simulations of the PET signal based on Geant4 and FLUKA Monte Carlo (MC) codes. To investigate the effects of patient motion, the lateral agreement between measured and simulated activity distributions for study cases were compared to the lateral conformity in head and neck tumor cases. Hounsfield Unit (HU) domains for different organs in the abdomen and pelvis were determined to investigate whether the HU mapping is sufficiently correlative to assign organ specific elemental compositions and washout characteristics in the MC simulations. Range comparisons between measurement and simulation at specific positions in the fall-off region of the activity (point wise verification) as well as a range analysis taking the entire fall-off region into account (shift verification) were performed. Analyses were based upon three pelvic and one spinal tumor cases. Results: Lateral blurring in abdominopelvic tumor sites ranged up to 25mm compared to prior documented conformity for head and neck tumor sites of up to 5mm. Lateral blurring is dominant in the anterior-posterior direction reflecting the breathing motion during the 30min of PET scan. The HU analysis showed overlapping HU domains, for example in bone marrow and soft tissue. This prevents an organ specific assignment based upon density and thus does not provide an accurate simulation of the total yield and spatial distribution of b+-activity. Mean range deviation between measurement and simulation in the prostate cases at the 20% and 50% positions in the last activity fall-off was 3.1mm and 5.4mm, respectively, and was 4.5mm by shift verification. For opposed treatment beams, point wise as well as shift range verification were found to be impractical. Conclusions: In abdominopelvic tumor sites, mm-accurate offline PET/CT range verification is not currently feasible primarily due to patient motion and the position of the distal beam edge in soft tissue. Author Disclosure: H.A. Shih, None; A. Knopf, None; K. Parodi, None; H. Paganetti, None; T. Bortfeld, None.
3052
Low-dose X-ray Fluoroscopy for Image Guided Radiation Therapy (IGRT)
X. Zhang, J. Wang, L. Zhu, L. Xing Stanford University School of Medicine, Stanford, CA Purpose/Objective(s): X-ray fluoroscopic imaging plays an important role in IGRT. There are growing interests in using X-ray fluoroscopy for the management of organ motion and gating in radiotherapy. However, long-time exposure of X-rays may lead to adverse health effect to the patients. In this work, we aim to improve the quality of low-dose X-ray fluoroscopic images using statistics-based restoration algorithm so that the patient fluoroscopy can be performed with much reduced radiation dose. Method/Materials: The noise in the low-dose fluoroscopy was suppressed by temporal plus spatial filtering. The correlation among neighboring frames was considered by the Karhunen-Loe`ve (KL) transform. After the KL transform, the selected neighboring frames of fluoroscopy was decomposed to un-correlated and ordered principal components; and each KL component was associated with an Eigen value. For each KL component, a penalized weighted least-squares (PWLS) objective function was constructed to restore the ideal image. The PWLS criterion reflects that the measurement with smaller signal-to-noise ratio (SNR) will contribute less to the objective function. The variance of each measurement was chosen as the weight since the measured signal can be modeled as Poisson noise and the measurement of larger variance has larger SNR for Poisson noise. The penalty was chosen as quadratic and the penalty parameter in each KL component was inverse proportional to its corresponding Eigen value. Smaller KL Eigen value is associated with the KL component of lower SNR, and a larger penalty parameter should be used for such KL component. The low-dose fluoroscopic images were acquired by a Varian Acuity simulator and a quality assurance phantom was used to evaluate the presented algorithm. Results: Low-dose fluoroscopic images were acquired with X-ray tube current of 10 mA and duration of X-ray pulse 2 ms. In the image restored by the proposed KL domain PWLS algorithm, noise is greatly suppressed while fine structures are well preserved. Comparison studies with traditional techniques, such as the mean and median filters, show that the proposed algorithm is advantageous in terms of resolution preservation. Conclusions: The proposed noise reduction algorithm can significantly improve the quality of low-dose X-ray fluoroscopic image and allows for dose reduction in X-ray fluoroscopy. Author Disclosure: X. Zhang, None; J. Wang, None; L. Zhu, None; L. Xing, None.
3053
Clinical use of Electromagnetic Guidance for Lung and Spine Radiation Therapy
T. R. Willoughby, A. P. Shah, A. R. Forbes, R. R. Manon, P. A. Kupelian, S. L. Meeks MD Anderson Cancer Center, Orlando, Orlando, FL Purpose/Objective(s): To describe the first clinical use of the Calypso 4D Localization System in regions outside of the prostate gland, specifically in spine and lung radiotherapy. Materials/Methods: The Calypso System (Calypso Medical, Seattle, WA) is approved for routine guidance for external beam radiotherapy of localized prostate cancers. However, other anatomic sites could benefit from similar guidance of high dose external