TY - JOUR
T1 - SU‐E‐J‐47
T2 - EPID Based Target Tracking During Intensity‐Modulated Radiation Therapy
AU - Ma, C.M.
AU - xu, Q.
AU - Tong, X.
AU - Lin, M.
AU - Chen, X.
AU - Fan, J.
AU - Chen, L.
PY - 2013/6
Y1 - 2013/6
N2 - Purpose: This work investigates the feasibility of using real‐time EPID images of the treatment fields acquired during IMRT for target localization and tracking. Methods: In this study, 35 patients treated with IMRT were retrospectively investigated. These patients were grouped as: prostate with lymph node (n=14), prostate without lymph node (n=17), and lung (n=4). For each patient, two to four fiducial markers were implanted inside the tumor. The DRR, which projects the patient anatomy and the fiducial marker at the EPID location, was reconstructed for each field. The MLC aperture of each control point was overlay on its corresponding DRR to evaluate the fractional time when the fiducial marker was seen on the EPID image. The probability of seeing at least one, two, three, or four fiducial markers during the treatment was recorded. Results: Our results show that for prostate IMRT patients without lymph nodes included in the target volume, the average probability of seeing at least one, two, three, and four fiducial markers during the treatment was 50% (35%–59%), 39% (23%– 51%), 24% (7%–38%), and 12% (4%–29%), respectively. For prostate IMRT patients with lymph nodes, the probability was 41% (24%–51%), 29% (12%– 42%), 15% (3%–24%), and 7% (4%–15%), respectively. For lung IMRT treatment, the average probability of seeing at least one fiducial marker was 34% (20%–52%). Conclusion: The continuous image acquisition from the EPID during IMRT treatment provides sufficient target movement information for real‐time target localization and intrafractional target motion correction for advanced radiotherapy treatments.
AB - Purpose: This work investigates the feasibility of using real‐time EPID images of the treatment fields acquired during IMRT for target localization and tracking. Methods: In this study, 35 patients treated with IMRT were retrospectively investigated. These patients were grouped as: prostate with lymph node (n=14), prostate without lymph node (n=17), and lung (n=4). For each patient, two to four fiducial markers were implanted inside the tumor. The DRR, which projects the patient anatomy and the fiducial marker at the EPID location, was reconstructed for each field. The MLC aperture of each control point was overlay on its corresponding DRR to evaluate the fractional time when the fiducial marker was seen on the EPID image. The probability of seeing at least one, two, three, or four fiducial markers during the treatment was recorded. Results: Our results show that for prostate IMRT patients without lymph nodes included in the target volume, the average probability of seeing at least one, two, three, and four fiducial markers during the treatment was 50% (35%–59%), 39% (23%– 51%), 24% (7%–38%), and 12% (4%–29%), respectively. For prostate IMRT patients with lymph nodes, the probability was 41% (24%–51%), 29% (12%– 42%), 15% (3%–24%), and 7% (4%–15%), respectively. For lung IMRT treatment, the average probability of seeing at least one fiducial marker was 34% (20%–52%). Conclusion: The continuous image acquisition from the EPID during IMRT treatment provides sufficient target movement information for real‐time target localization and intrafractional target motion correction for advanced radiotherapy treatments.
UR - http://www.scopus.com/inward/record.url?scp=85024781590&partnerID=8YFLogxK
U2 - 10.1118/1.4814259
DO - 10.1118/1.4814259
M3 - Article
AN - SCOPUS:85024781590
SN - 0094-2405
VL - 40
SP - 160
JO - Medical Physics
JF - Medical Physics
IS - 6
ER -