- Email: [email protected]
[P202] No improvement in kV-MV based setup for breast radiotherapy with optical surface guided correction of arm posture
158
Abstracts / Physica Medica 52 (2018) 99–187
Automatic-structure: voxels with unknown density (>3.00 g/cm3) are automatic drawn by SHDA-tool. We have assigned at this structure the Titanium Alloy density (4.42 g/cm3). No density are assigned in artifacts area in soft tissue regions (i.e. oral cavity mucosae); Manual structure: prosthetic teeth and soft tissue regions with artifacts are manually contouring and assigned Titanium Alloy and muscle-skeletal (1.05 g/cm3) density respectively.
At the end of optimization process, dose distributions were calculated on both CT set. For these we analyzed some clinical parameter from dose-volume histogram (DVH) of: PTV, mucosae, mucosae near implant interface, parotids and mandible. Results. Automatic-structure volume is bigger than manual one for 81% (medium volume differences 6.7 cc (0.5 19.9 cc)). No significant differences were observed between dose distributions in PTV coverture and mean dose of PTV, parotids and mucosae. The maximum doses for oral mucosae and mucosae near implants have differences <2.5 Gy for 40% cases; while for PTV and mandible the differences are <2.0 Gy for 25% patients. In the other cases there are not significant differences. Conclusions. The differences in dose distribution between the two methods, for each patient, are small and not clinically relevant. We decide to introduce the SHDA tool in our pianification instead of manual structures contouring. To improve the PTVs and OARs contouring we are evaluating the recently commercially algorithm for CT-image reconstruction. https://doi.org/10.1016/j.ejmp.2018.06.495
[P202] No improvement in kV-MV based setup for breast radiotherapy with optical surface guided correction of arm posture Susanne N. Bekke a, Faisal Mahmood b, Claus F. Behrens c,* a
Radiotherapy Research Unit, Department of Oncology, Herlev and Gentofte Hospital, University of Copenhagen, Center for Nuclear Technologies, Technical University of Denmark, Dtu Risø Campus, Department of Oncology, Herlev, Denmark b Research Unit for Oncology, Department of Clinical Research, University of Southern Denmark, Department of Oncology, Odense University Hospital, Department of Oncology, Odense, Denmark c Radiotherapy Research Unit, Herlev and Gentofte Hospital, University of Copenhagen, Oncology, Herlev, Denmark ⇑ Corresponding author. Purpose. Optical surface scanning can be used to correct the arm posture in breast radiotherapy. The purpose of this study was to investigate if such arm posture corrections can improve kV-MV based patient setup. Furthermore it was evaluated whether setup based on surface scanning alone would be sufficient. Methods. Twenty-one patients without and 19 patients with arm posture correction using an optical surface scanning system (Catalyst and Sentinel, from C-RAD Positioning AB, Uppsala, Sweden) were included. A surface scan was acquired, and patients were subsequently imaged using kV-MV imaging. Required translational shifts were applied to reach the treatment position. The patients were treated and then imaged using cone beam computed tomography CBCT. The setup errors in the two groups (with and without arm correction) were evaluated using the CBCT as ground truth. Based on alignment of the surfaces from treatment to the surface from the CT room the potential treatment positions, had setup been based on surface scanning alone, were calculated. Results. Arm posture correction was needed in 86% of the treatment fractions. Correcting the arm posture did not lead to any
statistical significant differences in setup errors compared to the group with no arm correction. If patient setup had been based solely on surface scanning statistical significant larger setup errors would have been a result. Conclusions. Using an optical surface scanning system for arm posture correction prior to kV-MV imaging in breast radiotherapy was not found to improve the kV-MV based setup. The study suggest that surface based setup can be used for initial setup, but target position should be verified using e.g. kV-MV imaging. https://doi.org/10.1016/j.ejmp.2018.06.496
[P203] WEPL-corrected margins for moving targets in carbon-ion gated treatments: Preliminary results Giorgia Meschini a,*, Alessandro Vai b, Marco Riboldi c, Andrea Pella d, Viviana Vitolo e, Donatella Caivano e, Mario Ciocca b, Guido Baroni a, Silvia Molinelli b a
Politecnico DI Milano, Department of Electronics, Information and Bioengineering, Milan, Italy b Centro Nazionale DI Adroterapia Oncologica, Medical Physics, Pavia, Italy c Ludwig-Maximilians-Universität, Chair of Experimental Physics – Medical Physics, München, Germany d Centro Nazionale DI Adroterapia Oncologica, Bioengineering Unit, Pavia, Italy e Centro Nazionale DI Adroterapia Oncologica, Medical Division, Pavia, Italy ⇑ Corresponding author. Purpose. To evaluate intra- and inter-fractional robustness of carbon-ion treatment plans based on water-equivalent path length (WEPL) corrected ITV margins for moving targets irradiated with gated pencil-beam scanning. Methods. As a preliminary study, three patients with thoracoabdominal targets were considered. Clinical plans were optimized with an Intensity Modulated Particle Therapy (IMPT) algorithm and the use of case-specific avoidance structures to prevent radiation pathways on regions subject to intra- and inter-fractional variations. Plans were optimized on the end-exhale CT of a 4DCT dataset using 2 fields and tested for consistency within the gating window by recalculation on the 30%-exhale and 30%-inhale CTs. Single Field Uniform Dose (SFUD) plans were optimized on beam-specific WEPLcorrected ITV (ITVWEPL ), which comprised the envelope of waterequivalent target positions within the ideal gating window. Differently from clinical plans, avoidance structures were not used in this approach. Clinical plans were taken as reference for comparison. Dose recalculations on the planning 4DCT and on synthetic 4DCT simulating irregular breathing allowed to evaluate intra-fractional robustness of the two approaches, whereas inter-fractional robustness was estimated on re-evaluation CTs (one per patient). Differences in the conformity index CI ¼ V95% =Vtarget , target coverage D95% and homogeneity index HI = (D2% D98% )/D50% between the end-exhale plan and the recalculated dose distributions were computed. Doses to the involved organs at risk (OARs) were evaluated as well. Results. The CI presented slightly higher median intra-/interfractional variations and absolute values in the ITVWEPL approach (13%–1.47) with respect to reference plans (6%–1.36). D95% and HI showed lower median intra-fractional variations for the ITVWEPL approach (0.1% and 35%) compared to reference plans (3.4% and 46%). In case of inter-fractional changes, the use of ITVWEPL led to higher median variations but comparable absolute values for D95% (3.8% vs 2.2%) and for HI (144% vs 20%, absolute median values: 0.11 vs 0.14). The two methods performed similarly in OARs.
Abstracts / Physica Medica 52 (2018) 99–187
Automatic-structure: voxels with unknown density (>3.00 g/cm3) are automatic drawn by SHDA-tool. We have assigned at this structure the Titanium Alloy density (4.42 g/cm3). No density are assigned in artifacts area in soft tissue regions (i.e. oral cavity mucosae); Manual structure: prosthetic teeth and soft tissue regions with artifacts are manually contouring and assigned Titanium Alloy and muscle-skeletal (1.05 g/cm3) density respectively.
At the end of optimization process, dose distributions were calculated on both CT set. For these we analyzed some clinical parameter from dose-volume histogram (DVH) of: PTV, mucosae, mucosae near implant interface, parotids and mandible. Results. Automatic-structure volume is bigger than manual one for 81% (medium volume differences 6.7 cc (0.5 19.9 cc)). No significant differences were observed between dose distributions in PTV coverture and mean dose of PTV, parotids and mucosae. The maximum doses for oral mucosae and mucosae near implants have differences <2.5 Gy for 40% cases; while for PTV and mandible the differences are <2.0 Gy for 25% patients. In the other cases there are not significant differences. Conclusions. The differences in dose distribution between the two methods, for each patient, are small and not clinically relevant. We decide to introduce the SHDA tool in our pianification instead of manual structures contouring. To improve the PTVs and OARs contouring we are evaluating the recently commercially algorithm for CT-image reconstruction. https://doi.org/10.1016/j.ejmp.2018.06.495
[P202] No improvement in kV-MV based setup for breast radiotherapy with optical surface guided correction of arm posture Susanne N. Bekke a, Faisal Mahmood b, Claus F. Behrens c,* a
Radiotherapy Research Unit, Department of Oncology, Herlev and Gentofte Hospital, University of Copenhagen, Center for Nuclear Technologies, Technical University of Denmark, Dtu Risø Campus, Department of Oncology, Herlev, Denmark b Research Unit for Oncology, Department of Clinical Research, University of Southern Denmark, Department of Oncology, Odense University Hospital, Department of Oncology, Odense, Denmark c Radiotherapy Research Unit, Herlev and Gentofte Hospital, University of Copenhagen, Oncology, Herlev, Denmark ⇑ Corresponding author. Purpose. Optical surface scanning can be used to correct the arm posture in breast radiotherapy. The purpose of this study was to investigate if such arm posture corrections can improve kV-MV based patient setup. Furthermore it was evaluated whether setup based on surface scanning alone would be sufficient. Methods. Twenty-one patients without and 19 patients with arm posture correction using an optical surface scanning system (Catalyst and Sentinel, from C-RAD Positioning AB, Uppsala, Sweden) were included. A surface scan was acquired, and patients were subsequently imaged using kV-MV imaging. Required translational shifts were applied to reach the treatment position. The patients were treated and then imaged using cone beam computed tomography CBCT. The setup errors in the two groups (with and without arm correction) were evaluated using the CBCT as ground truth. Based on alignment of the surfaces from treatment to the surface from the CT room the potential treatment positions, had setup been based on surface scanning alone, were calculated. Results. Arm posture correction was needed in 86% of the treatment fractions. Correcting the arm posture did not lead to any
statistical significant differences in setup errors compared to the group with no arm correction. If patient setup had been based solely on surface scanning statistical significant larger setup errors would have been a result. Conclusions. Using an optical surface scanning system for arm posture correction prior to kV-MV imaging in breast radiotherapy was not found to improve the kV-MV based setup. The study suggest that surface based setup can be used for initial setup, but target position should be verified using e.g. kV-MV imaging. https://doi.org/10.1016/j.ejmp.2018.06.496
[P203] WEPL-corrected margins for moving targets in carbon-ion gated treatments: Preliminary results Giorgia Meschini a,*, Alessandro Vai b, Marco Riboldi c, Andrea Pella d, Viviana Vitolo e, Donatella Caivano e, Mario Ciocca b, Guido Baroni a, Silvia Molinelli b a
Politecnico DI Milano, Department of Electronics, Information and Bioengineering, Milan, Italy b Centro Nazionale DI Adroterapia Oncologica, Medical Physics, Pavia, Italy c Ludwig-Maximilians-Universität, Chair of Experimental Physics – Medical Physics, München, Germany d Centro Nazionale DI Adroterapia Oncologica, Bioengineering Unit, Pavia, Italy e Centro Nazionale DI Adroterapia Oncologica, Medical Division, Pavia, Italy ⇑ Corresponding author. Purpose. To evaluate intra- and inter-fractional robustness of carbon-ion treatment plans based on water-equivalent path length (WEPL) corrected ITV margins for moving targets irradiated with gated pencil-beam scanning. Methods. As a preliminary study, three patients with thoracoabdominal targets were considered. Clinical plans were optimized with an Intensity Modulated Particle Therapy (IMPT) algorithm and the use of case-specific avoidance structures to prevent radiation pathways on regions subject to intra- and inter-fractional variations. Plans were optimized on the end-exhale CT of a 4DCT dataset using 2 fields and tested for consistency within the gating window by recalculation on the 30%-exhale and 30%-inhale CTs. Single Field Uniform Dose (SFUD) plans were optimized on beam-specific WEPLcorrected ITV (ITVWEPL ), which comprised the envelope of waterequivalent target positions within the ideal gating window. Differently from clinical plans, avoidance structures were not used in this approach. Clinical plans were taken as reference for comparison. Dose recalculations on the planning 4DCT and on synthetic 4DCT simulating irregular breathing allowed to evaluate intra-fractional robustness of the two approaches, whereas inter-fractional robustness was estimated on re-evaluation CTs (one per patient). Differences in the conformity index CI ¼ V95% =Vtarget , target coverage D95% and homogeneity index HI = (D2% D98% )/D50% between the end-exhale plan and the recalculated dose distributions were computed. Doses to the involved organs at risk (OARs) were evaluated as well. Results. The CI presented slightly higher median intra-/interfractional variations and absolute values in the ITVWEPL approach (13%–1.47) with respect to reference plans (6%–1.36). D95% and HI showed lower median intra-fractional variations for the ITVWEPL approach (0.1% and 35%) compared to reference plans (3.4% and 46%). In case of inter-fractional changes, the use of ITVWEPL led to higher median variations but comparable absolute values for D95% (3.8% vs 2.2%) and for HI (144% vs 20%, absolute median values: 0.11 vs 0.14). The two methods performed similarly in OARs.