Purpose: To develop and evaluate the feasibility of deep learning approaches for MR-based treatme... more Purpose: To develop and evaluate the feasibility of deep learning approaches for MR-based treatment planning (deepMTP) in brain tumor radiation therapy. Methods and materials: A treatment planning pipeline was constructed using a deep learning approach to generate continuously valued pseudo CT images from MR images. A deep convolutional neural network was designed to identify tissue features in volumetric head MR images training with co-registered kVCT images. A set of 40 retrospective 3D T1-weighted head images was utilized to train the model, and evaluated in 10 clinical cases with brain metastases by comparing treatment plans using deep learning generated pseudo CT and using an acquired planning kVCT. Paired-sample Wilcoxon signed rank sum tests were used for statistical analysis to compare dosimetric parameters of plans made with pseudo CT images generated from deepMTP to those made with kVCT-based clinical treatment plan (CTTP). Results: deepMTP provides an accurate pseudo CT with Dice coefficients for air: 0.95 ± 0.01, soft tissue: 0.94 ± 0.02, and bone: 0.85 ± 0.02 and a mean absolute error of 75 ± 23 HU compared with acquired kVCTs. The absolute percentage differences of dosimetric parameters between deepMTP and CTTP was 0.24% ± 0.46% for planning target volume (PTV) volume, 1.39% ± 1.31% for maximum dose and 0.27% ± 0.79% for the PTV receiving 95% of the prescribed dose (V95). Furthermore, no significant difference was found for PTV volume (P = 0.50), the maximum dose (P = 0.83) and V95 (P = 0.19) between deepMTP and CTTP. Conclusions: We have developed an automated approach (deepMTP) that allows generation of a continuously valued pseudo CT from a single high-resolution 3D MR image and evaluated it in partial brain tumor treatment planning. The deepMTP provided dose distribution with no significant difference relative to a kVCT-based standard volumetric modulated arc therapy plans. KEY WORDS brain tumor, convolutional neural network, deep learning, MRI, MR-only treatment planning,
The purpose of this study was to evaluate helical tomotherapy dosimetry in postmastectomy patient... more The purpose of this study was to evaluate helical tomotherapy dosimetry in postmastectomy patients undergoing treatment for chest wall and positive nodal regions with simultaneous integrated boost (SIB) in the scar region using strip bolus. Six postmastectomy patients were scanned with a 5-mm-thick strip bolus covering the scar planning target volume (PTV) plus 2-cm margin. For all 6 cases, the chest wall received a total cumulative dose of 49.3-50.4 Gy with daily fraction size of 1.7-2.0 Gy. Total dose to the scar PTV was prescribed to 58.0 -60.2 Gy at 2.0 -2.5 Gy per fraction. The supraclavicular PTV and mammary nodal PTV received 1.7-1.9 dose per fraction. Two plans (with and without bolus) were generated for all 6 cases. To generate no-bolus plans, strip bolus was contoured and overrode to air density before planning. The setup reproducibility and delivered dose accuracy were evaluated for all 6 cases. Dose-volume histograms were used to evaluate dose-volume coverage of targets and critical structures. We observed reduced air cavities with the strip bolus setup compared with what we normally see with the full bolus. The thermoluminescence dosimeters (TLD) in vivo dosimetry confirmed accurate dose delivery beneath the bolus. The verification plans performed on the first day megavoltage computed tomography (MVCT) image verified that the daily setup and overall dose delivery was within 2% accuracy compared with the planned dose. The hotspot of the scar PTV in no-bolus plans was 111.4% of the prescribed dose averaged over 6 cases compared with 106.6% with strip bolus. With a strip bolus only covering the postmastectomy scar region, we observed increased dose uniformity to the scar PTV, higher setup reproducibility, and accurate dose delivered beneath the bolus. This study demonstrates the feasibility of using a strip bolus over the scar using tomotherapy for SIB dosimetry in postmastectomy treatments.
This study highlights the use of adaptive planning to accommodate testicular shielding in helical... more This study highlights the use of adaptive planning to accommodate testicular shielding in helical tomotherapy for malignancies of the proximal thigh. Two cases of young men with large soft tissue sarcomas of the proximal thigh are presented. After multidisciplinary evaluation, preoperative radiation therapy was recommended. Both patients were referred for sperm banking and lead shields were used to minimize testicular dose during radiation therapy. To minimize imaging artifacts, kilovoltage CT (kVCT) treatment planning was conducted without shielding. Generous hypothetical contours were generated on each “planning scan” to estimate the location of the lead shield and generate a directionally blocked helical tomotherapy plan. To ensure the accuracy of each plan, megavoltage fan-beam CT (MVCT) scans were obtained at the first treatment and adaptive planning was performed to account for lead shield placement. Two important regions of interest in these cases were femurs and femoral heads. During adaptive planning for the first patient, it was observed that the virtual lead shield contour on kVCT planning images was significantly larger than the actual lead shield used for treatment. However, for the second patient, it was noted that the size of the virtual lead shield contoured on the kVCT image was significantly smaller than the actual shield size. Thus, new adaptive plans based on MVCT images were generated and used for treatment. The planning target volume was underdosed up to 2% and had higher maximum doses without adaptive planning. In conclusion, the treatment of the upper thigh, particularly in young men, presents several clinical challenges, including preservation of gonadal function. In such circumstances, adaptive planning using MVCT can ensure accurate dose delivery even in the presence of high-density testicular shields.
Purpose: To develop and evaluate the feasibility of deep learning approaches for MR-based treatme... more Purpose: To develop and evaluate the feasibility of deep learning approaches for MR-based treatment planning (deepMTP) in brain tumor radiation therapy. Methods and materials: A treatment planning pipeline was constructed using a deep learning approach to generate continuously valued pseudo CT images from MR images. A deep convolutional neural network was designed to identify tissue features in volumetric head MR images training with co-registered kVCT images. A set of 40 retrospective 3D T1-weighted head images was utilized to train the model, and evaluated in 10 clinical cases with brain metastases by comparing treatment plans using deep learning generated pseudo CT and using an acquired planning kVCT. Paired-sample Wilcoxon signed rank sum tests were used for statistical analysis to compare dosimetric parameters of plans made with pseudo CT images generated from deepMTP to those made with kVCT-based clinical treatment plan (CTTP). Results: deepMTP provides an accurate pseudo CT with Dice coefficients for air: 0.95 ± 0.01, soft tissue: 0.94 ± 0.02, and bone: 0.85 ± 0.02 and a mean absolute error of 75 ± 23 HU compared with acquired kVCTs. The absolute percentage differences of dosimetric parameters between deepMTP and CTTP was 0.24% ± 0.46% for planning target volume (PTV) volume, 1.39% ± 1.31% for maximum dose and 0.27% ± 0.79% for the PTV receiving 95% of the prescribed dose (V95). Furthermore, no significant difference was found for PTV volume (P = 0.50), the maximum dose (P = 0.83) and V95 (P = 0.19) between deepMTP and CTTP. Conclusions: We have developed an automated approach (deepMTP) that allows generation of a continuously valued pseudo CT from a single high-resolution 3D MR image and evaluated it in partial brain tumor treatment planning. The deepMTP provided dose distribution with no significant difference relative to a kVCT-based standard volumetric modulated arc therapy plans. KEY WORDS brain tumor, convolutional neural network, deep learning, MRI, MR-only treatment planning,
The purpose of this study was to evaluate helical tomotherapy dosimetry in postmastectomy patient... more The purpose of this study was to evaluate helical tomotherapy dosimetry in postmastectomy patients undergoing treatment for chest wall and positive nodal regions with simultaneous integrated boost (SIB) in the scar region using strip bolus. Six postmastectomy patients were scanned with a 5-mm-thick strip bolus covering the scar planning target volume (PTV) plus 2-cm margin. For all 6 cases, the chest wall received a total cumulative dose of 49.3-50.4 Gy with daily fraction size of 1.7-2.0 Gy. Total dose to the scar PTV was prescribed to 58.0 -60.2 Gy at 2.0 -2.5 Gy per fraction. The supraclavicular PTV and mammary nodal PTV received 1.7-1.9 dose per fraction. Two plans (with and without bolus) were generated for all 6 cases. To generate no-bolus plans, strip bolus was contoured and overrode to air density before planning. The setup reproducibility and delivered dose accuracy were evaluated for all 6 cases. Dose-volume histograms were used to evaluate dose-volume coverage of targets and critical structures. We observed reduced air cavities with the strip bolus setup compared with what we normally see with the full bolus. The thermoluminescence dosimeters (TLD) in vivo dosimetry confirmed accurate dose delivery beneath the bolus. The verification plans performed on the first day megavoltage computed tomography (MVCT) image verified that the daily setup and overall dose delivery was within 2% accuracy compared with the planned dose. The hotspot of the scar PTV in no-bolus plans was 111.4% of the prescribed dose averaged over 6 cases compared with 106.6% with strip bolus. With a strip bolus only covering the postmastectomy scar region, we observed increased dose uniformity to the scar PTV, higher setup reproducibility, and accurate dose delivered beneath the bolus. This study demonstrates the feasibility of using a strip bolus over the scar using tomotherapy for SIB dosimetry in postmastectomy treatments.
This study highlights the use of adaptive planning to accommodate testicular shielding in helical... more This study highlights the use of adaptive planning to accommodate testicular shielding in helical tomotherapy for malignancies of the proximal thigh. Two cases of young men with large soft tissue sarcomas of the proximal thigh are presented. After multidisciplinary evaluation, preoperative radiation therapy was recommended. Both patients were referred for sperm banking and lead shields were used to minimize testicular dose during radiation therapy. To minimize imaging artifacts, kilovoltage CT (kVCT) treatment planning was conducted without shielding. Generous hypothetical contours were generated on each “planning scan” to estimate the location of the lead shield and generate a directionally blocked helical tomotherapy plan. To ensure the accuracy of each plan, megavoltage fan-beam CT (MVCT) scans were obtained at the first treatment and adaptive planning was performed to account for lead shield placement. Two important regions of interest in these cases were femurs and femoral heads. During adaptive planning for the first patient, it was observed that the virtual lead shield contour on kVCT planning images was significantly larger than the actual lead shield used for treatment. However, for the second patient, it was noted that the size of the virtual lead shield contoured on the kVCT image was significantly smaller than the actual shield size. Thus, new adaptive plans based on MVCT images were generated and used for treatment. The planning target volume was underdosed up to 2% and had higher maximum doses without adaptive planning. In conclusion, the treatment of the upper thigh, particularly in young men, presents several clinical challenges, including preservation of gonadal function. In such circumstances, adaptive planning using MVCT can ensure accurate dose delivery even in the presence of high-density testicular shields.
Uploads
Papers by Poonam Yadav
tomotherapy for malignancies of the proximal thigh. Two cases of young men with large soft tissue
sarcomas of the proximal thigh are presented. After multidisciplinary evaluation, preoperative radiation
therapy was recommended. Both patients were referred for sperm banking and lead shields were used to
minimize testicular dose during radiation therapy. To minimize imaging artifacts, kilovoltage CT (kVCT)
treatment planning was conducted without shielding. Generous hypothetical contours were generated on
each “planning scan” to estimate the location of the lead shield and generate a directionally blocked helical
tomotherapy plan. To ensure the accuracy of each plan, megavoltage fan-beam CT (MVCT) scans were
obtained at the first treatment and adaptive planning was performed to account for lead shield placement.
Two important regions of interest in these cases were femurs and femoral heads. During adaptive planning
for the first patient, it was observed that the virtual lead shield contour on kVCT planning images was
significantly larger than the actual lead shield used for treatment. However, for the second patient, it was
noted that the size of the virtual lead shield contoured on the kVCT image was significantly smaller than
the actual shield size. Thus, new adaptive plans based on MVCT images were generated and used for
treatment. The planning target volume was underdosed up to 2% and had higher maximum doses without
adaptive planning. In conclusion, the treatment of the upper thigh, particularly in young men, presents
several clinical challenges, including preservation of gonadal function. In such circumstances, adaptive
planning using MVCT can ensure accurate dose delivery even in the presence of high-density testicular
shields.
tomotherapy for malignancies of the proximal thigh. Two cases of young men with large soft tissue
sarcomas of the proximal thigh are presented. After multidisciplinary evaluation, preoperative radiation
therapy was recommended. Both patients were referred for sperm banking and lead shields were used to
minimize testicular dose during radiation therapy. To minimize imaging artifacts, kilovoltage CT (kVCT)
treatment planning was conducted without shielding. Generous hypothetical contours were generated on
each “planning scan” to estimate the location of the lead shield and generate a directionally blocked helical
tomotherapy plan. To ensure the accuracy of each plan, megavoltage fan-beam CT (MVCT) scans were
obtained at the first treatment and adaptive planning was performed to account for lead shield placement.
Two important regions of interest in these cases were femurs and femoral heads. During adaptive planning
for the first patient, it was observed that the virtual lead shield contour on kVCT planning images was
significantly larger than the actual lead shield used for treatment. However, for the second patient, it was
noted that the size of the virtual lead shield contoured on the kVCT image was significantly smaller than
the actual shield size. Thus, new adaptive plans based on MVCT images were generated and used for
treatment. The planning target volume was underdosed up to 2% and had higher maximum doses without
adaptive planning. In conclusion, the treatment of the upper thigh, particularly in young men, presents
several clinical challenges, including preservation of gonadal function. In such circumstances, adaptive
planning using MVCT can ensure accurate dose delivery even in the presence of high-density testicular
shields.