ABSTRACT
Objective:To assess the feasibility of deformable image registration (DIR) in evaluating cumulative dose distribution of bladder and rectum of cervical cancer patients during helical tomotherapy (HT)-based intensity-modulated radiotherapy (IMRT) and high-dose-rate(HDR) brachytherapy.Methods:Clinical data of 18 patients were retrospectively analyzed. Cumulative bladder/rectum D 2cm 3 and high-risk clinical target volume (HR-CTV) D 90% parameters were calculated and compared to two direct parameter-adding methods with two registration-adding methods. Equivalent uniform dose (EUD group) and overlapping high dose (OHD group) methods were employed as parameter-adding methods. The registration-adding methods including rigid image registration (RIR group) and deformable image registration (DIR group) were adopted based on a commercial image registration software (MIM Maestro ?). The dice similarity coefficient (DSC) and mean distance to agreement (MDA) were measured to assess the accuracy of RIR and DIR. Results:In the EUD, OHD, RIR and DIR groups, the cumulative doses of bladder/rectum D 2cm 3 and HR-CTV D 90% were (80.11±3.59) Gy (EQD 2Gy), (82.23±3.46) Gy (EQD 2Gy), (80.99±6.01) Gy (EQD 2Gy) and (81.19±3.11) Gy (EQD 2Gy)( P=0.516); (72.90±3.58) Gy (EQD 2Gy), (73.83±4.28) Gy (EQD 2Gy), (72.45±6.05) Gy (EQD 2Gy) and (71.98±2.89) Gy (EQD 2Gy)( P=0.625), and (85.51±2.91) Gy (EQD 2Gy), (87.65±3.46) Gy (EQD 2Gy), (81.53±3.63) Gy (EQD 2Gy) and (85.81±3.30) Gy (EQD 2Gy)( P<0.001), respectively. The mean DSC of the bladder, rectum and HR-CTV were 0.69, 0.65 and 0.63 with RIR; and 0.85, 0.81 and 0.78 with DIR ( P<0.001), respectively. In DIR, the average MDA of bladder, rectum, and HR-CTV were 2.88, 2.48 and 2.66 mm, respectively. Conclusions:The cumulative DVH parameters among 4 groups show no significant difference in the bladder/rectum D 2cm 3/D 0.2cm 3. Since the DIR group achieves satisfactory volume matching of greater than 0.8 with DSC analysis, it can yield acceptable results for clinical application between HT IMRT and HDR BT for cervical cancer.
ABSTRACT
Objective:To study the differences in the cumulative dose between deformable image registration (DIR) and simple dose-volume histogram (DVH) summation in the fractionated brachytherapy of cervical cancer, and to analyze the feasibility of the application of DIR in the dosimetry assessment of targets and organs-at-risk (OARs) in the brachytherapy.Methods:A retrospective analysis was conducted for 13 cases with primary cervical cancer treated with four fractions of interstitial brachytherapy guided by CT images. The four CT images of each cases were registered using an intensity-based DIR. Then, the cumulative doses (the D2 cm 3, D1 cm 3, and D0.1 cm 3 of the bladder, rectum, intestine, and colon and the D90for targets) after DIR were calculated and compared to those obtained using simple DVH summation. Afterward, the correlation between the dose difference and dice similarity coefficient (DSC) was analyzed. With the dose difference (the remaining dose of OARs caused by the DIR) as limits, a new plan was made for the latest CT to calculate the dose increase to targets. Results:Compared to simple DVH summation, DIR allowed the cumulative doses of the D2 cm 3 and D1 cm 3 of bladder to be decreased by (2.47±1.92) and (2.82±2.73) Gy, respectively on average ( t=-3.65, -2.93, P < 0.05), those of the D2 cm 3, D1 cm 3, and D0.1 cm 3 of rectum to be decreased by (2.05 ± 1.61) Gy, (1.51 ± 1.58), and (3.21 ± 2.50) Gy, respectively on average ( t=-4.02, -3.02, -4.06, P < 0.05), and those of the D2 cm 3, D1 cm 3, and D0.1 cm 3 to be decreased by (1.42 ± 0.99), (1.55 ± 1.28) Gy, and (2.43 ± 1.95) Gy, respectively on average ( t=-3.52, -2.96, -3.06, P < 0.05). There was no significant statistical difference in the D90 of targets, the D0.1 cm 3 of the bladder, and the D2 cm 3, D1 cm 3, D0.1 cm 3 of the colon ( P > 0.05) between both methods, and there was no distinct correlation between DSC and dose difference ( P > 0.05). The DIR increased the dose to targets, with a median value of 150 cGy. However, the accuracy of the DIR should be improved. Conclusions:In clinical practice of multiple fractions of brachytherapy for cervical cancer, it′s still recommended to adopt the simple dose summation method to assess the doses to targets and OARs.
ABSTRACT
Objective@#To explore the differences and correlation between the target volumes based on deformation registration (DIR) using preoperative prone diagnostic magnetic resonance (MR) imaging and postoperative prone computed tomography (CT) simulation imaging for patients undergoing breast-conserving surgery (BCS).@*Methods@#Eighteen breast cancer patients suitable for external-beam partial breast irradiation (EB-PBI) after BCS were enrolled. Preoperative prone diagnostic MR and postoperative prone CT scan sets were acquired during free breathing for all patients. The gross tumor volume (GTV) delineated on the preoperative diagnostic MR images was defined as GTVMRI, the clinical target volumes (CTVMRI+ 1 and CTVMRI+ 2)were defined as 10 and 20 mm margins around the GTVMRI, and the planning target volume (PTVMRI+ 1 and PTVMRI+ 2) were defined as 15 and 25 mm margins around the GTVMRI, respectively. Tumor bed (TB) delineated on the postoperative prone CT simulation images acquired during free breathing was defined as GTVTB, CTV and PTV were defined as 10 and 15 mm margins around the GTVTB, respectively. The target volume of the whole breast contoured on the MR and CT images were defined as CTVBreast-MRI and CTVBreast-CI, respectively. The MR and CT images were registered deformably in MIM software system.@*Results@#The GTVTB, CTVTB and PTVTB were significantly greater than GTVMRI, GTVMRI+ 1 and PTVMRI+ 1, respectively (Z=-3.593, -3.593, -2.983, P<0.05). Meanwhile, the CTVTB and PTVTB were significantly less than the CTVMRI+ 2 and PTVMRI+ 2, respectively(Z=-2.722, -2.853, P<0.05). The conformal index (CI) and degree of inclusion (DI) of GTVTB-GTVMRI, GTVTB-CTVMRI+ 1, CTVTB-GTVMRI and CTVTB-GTVMRI+ 1 based on center-coincidence of the compared targets were better than those based on DIR of the thorax(Z=-3.724、-3.724、-2.591、-3.593, P<0.05; Z=-3.724、-3.724、-3.201、-3.724, P<0.05).@*Conclusions@#For the patients enrolled for prone EB-PBI, target volumes delineated on the preoperative prone MR images were significantly smaller compared to that on the postoperative prone CT images, but a statistically significant positive correlation was found between the MR and CT target volumes. There were still relatively poor spatial overlap whether for the whole breast or the targets between the preoperative prone diagnostic MR images and the postoperative prone simulation CT images based on DIR. Therefore, it is infeasible to guide postoperative EB-PBI target delineation using the preoperative prone diagnostic MR images.
ABSTRACT
Objective To evaluate the effect of different methods of deformable image registration on the dose evaluation in adaptive radiotherapy for lung cancer. Methods By using Raystation Ver4. 5 treatment planning system platform, two algorithms ( Hybrid-and Biomechanics-based deformable image registration) and two orders (CT images before and during radiotherapy as reference images) were adopted. Four deformable image registration methods were utilized to calculate the accumulative dose. Eleven patients of lung cancer received adaptive radiotherapy for 35. 0-61. 6 Gy were recruited. The mean doses of lung,heart and GTV and the D98 and D2 of GTV were statistically compared using four methods. Results With the four deformable image registration methods, the standard deviation of the mean lung dose of 11 lung cancer patients was ranged from 0. 07 to 0. 70 Gy,0. 01 to 0. 79 Gy for the mean heart dose,0. 01 to 2. 23 Gy for the mean GTV dose,0. 02 to 6. 51 Gy for the D98 of GTV and 0. 01 to 0. 97 Gy for the D2 of GTV,respectively. Conclusion The selection of deformable image registration method causes uncertainty to the calculation of accumulative dose during adaptive radiotherapy for lung cancer.
ABSTRACT
Objective To compare the geometric differences of gross tumor volumes(GTV)and displacements of selected clips propagated by rigid image registration(RIR)and deformable image registration (DIR)at end-inhale phase(CT0)and end-exhale phase(CT50)based on four-dimensional computed tomography(4DCT)of the whole breast after breast-conserving surgery(BCS). Methods Forty-four patients who underwent 4DCT simulation scans after BCS were selected. The GTV and displacements of selected metal clips at CT0and CT50were manually delineated by the same radiotherapy physician. Subsequently,the GTV and displacements of selected clips from CT0images were transformed and propagated to CT50images using RIR and DIR.The geometric differences of GTV and displacements of surgical clips from DIR were compared with those from RIR based on the dice similarity coefficient(DSC)and the displacements of the center of mass(COM)in the three-dimensional(3D)directions. Results The mean DSC was 0.86± 0.04 for RIR and 0.87± 0.04 for DIR(P=0.000).The displacements of COM in 3D directions from RIR were significantly greater than those from DIR(1.22 mm vs. 1.10 mm,P=0.000).In the anterior-posterior direction,the displacements from RIR were significantly greater than those from DIR for both GTV and selected clips(P=0.000).However,in the left-right and superior-inferior directions,there were no significant differences in displacements between RIR and DIR for both GTV and the selected clips(all P>0.05). Conclusions DIR can improve the overlap for GTV registration from 4DCT scans at CT0and CT50.Furthermore,DIR is superior to RIR in reflecting GTV and the displacements of selected clips in anterior-posterior direction induced by respiratory movement.
ABSTRACT
The number of imaging data sets has significantly increased during radiation treatment after introducing a diverse range of advanced techniques into the field of radiation oncology. As a consequence, there have been many studies proposing meaningful applications of imaging data set use. These applications commonly require a method to align the data sets at a reference. Deformable image registration (DIR) is a process which satisfies this requirement by locally registering image data sets into a reference image set. DIR identifies the spatial correspondence in order to minimize the differences between two or among multiple sets of images. This article describes clinical applications, validation, and algorithms of DIR techniques. Applications of DIR in radiation treatment include dose accumulation, mathematical modeling, automatic segmentation, and functional imaging. Validation methods discussed are based on anatomical landmarks, physical phantoms, digital phantoms, and per application purpose. DIR algorithms are also briefly reviewed with respect to two algorithmic components: similarity index and deformation models.
Subject(s)
Dataset , Methods , Models, Theoretical , Radiation OncologyABSTRACT
Objective To use helical tomotherapy ( HT ) for determining the difference between actual doses and planning doses to the target volume and organs at risk ( OARs ) in patients with nasopharyngeal carcinoma receiving radiotherapy, and to provide guidance for the clinical treatment. Methods Localization and delineation of the target volume and OARs were performed by computed tomography ( CT) in 21 patients with nasopharyngeal carcinoma receiving radical radiotherapy using HT. All patients underwent megavoltage CT ( MVCT) scan prior to treatment. The obtained MVCT images were used for dose reconstruction in the adaptive module of HT, in which the actual dose was obtained and the non?image?guided dose was simulated. Each single dose distribution and the corresponding CT image were sent to software MIM6. 0 for superimposition, and the overall dose was obtained. The initial plan, image?guided plan, and non?image?guided plan were named Plan?1, 2, and 3, respectively. The dose distribution in the target volume and OARs was compared between the three plans with t ? test or wilcoxon test . Results Compared with those in Plan?1, the D98 values for the planning gross tumor volume ( PGTV) and planning target volume ( PTV) in Plan?2 were significantly reduced by 1. 16% and 2. 3%, respectively ( P=0. 025;P=0. 043);the volumes of the left and right parotids in Plan?2 were significantly reduced by 46. 0% and 46. 5% on average, respectively ( P=0. 000);the distances between the midline and the center?of?mass for left and right parotids were significantly reduced by 6. 9% and 6. 5%, respectively ( P=0. 000);the V26 and Dmean for both parotid glands were significantly elevated by 32. 8% and 25. 2%, respectively ( P=0. 000) . Compared with those in Plan?1, the D98 values for PGTV, PTV?1, and PTV?2 in Plan?3 were significantly reduced by 2. 0%, 1. 9%, and 2. 4%, respectively ( P=0. 001;P=0. 007;P=0. 036);the V26 and Dmean for both parotid glands in Plan?3 were significantly elevated by 33. 6% and 25. 3%, respectively ( P=0. 000);Dmax to the spinal cord was significantly increased by 6. 9%( P=0. 005) . There was no significant difference in D2 to the spinal cord between Plan?2 and Plan?1( P=0. 392) . Conclusions The doses to both parotid glands increase during HT for nasopharyngeal carcinoma, which is closely associated with the shift of the parotid glands toward the midline. Image?guided radiotherapy does not enhance the dose to the target volume, but reduces the dose to the spinal cord.
ABSTRACT
PURPOSE: This study aimed to evaluate whether the deformable image registration (DIR) method is clinically applicable to the safe delivery of re-irradiation in hepatocellular carcinoma (HCC). MATERIALS AND METHODS: Between August 2010 and March 2012, 12 eligible HCC patients received re-irradiation using helical tomotherapy. The median total prescribed radiation doses at first irradiation and re-irradiation were 50 Gy (range, 36-60 Gy) and 50 Gy (range, 36-58.42 Gy), respectively. Most re-irradiation therapies (11 of 12) were administered to previously irradiated or marginal areas. Dose summation results were reproduced using DIR by rigid and deformable registration methods, and doses of organs-at-risk (OARs) were evaluated. Treatment outcomes were also assessed. RESULTS: Thirty-six dose summation indices were obtained for three OARs (bowel, duodenum, and stomach doses in each patient). There was no statistical difference between the two different types of DIR methods (rigid and deformable) in terms of calculated summation operatorD (0.1 cc, 1 cc, 2 cc, and max) in each OAR. The median total mean remaining liver doses (M(RLD)) in rigid- and deformable-type registration were not statistically different for all cohorts (p=0.248), although a large difference in M(RLD) was observed when there was a significant difference in spatial liver volume change between radiation intervals. One duodenal ulcer perforation developed 20 months after re-irradiation. CONCLUSION: Although current dose summation algorithms and uncertainties do not warrant accurate dosimetric results, OARs-based DIR dose summation can be usefully utilized in the re-irradiation of HCC. Appropriate cohort selection, watchful interpretation, and selective use of DIR methods are crucial to enhance the radio-therapeutic ratio.
Subject(s)
Adult , Aged , Female , Humans , Male , Middle Aged , Algorithms , Carcinoma, Hepatocellular/radiotherapy , Liver Neoplasms/radiotherapy , Organs at Risk/radiation effects , Radiation Dosage , Radiometry/methods , Radiotherapy/methods , Radiotherapy Dosage , Radiotherapy Planning, Computer-Assisted , Radiotherapy, Intensity-Modulated , Re-Irradiation , Tomography, X-Ray Computed/methods , Treatment OutcomeABSTRACT
Objective To study a novel method for the high-dose-rate brachytherapy (HDR) CT image to the intensity modulated radiation therapy (IMRT) CT image deformable image registration and dose accumulation.Methods The applicator in the HDR CT image is first segmented and removed,then a deflation step is performed on the applicator-free HDR CT image by solving the Navier-Stokes equation.Demons algorithm is utilized to register the deflated HDR CT image to the IMRT CT image,along with the HDR dose.The deformed HDR dose is then added on the IMRT dose and yield the final accumulated dose.Results The HDR CT image and IMRT CT image,as well as the corresponding dose distribution,from five cervical cancer patients are used for evaluation of the proposed algorithm,the results show that the proposed method can effectively get rid of the influence of the applicator and produce an accurate accumulated dose.Conclusions Dose accumulation and supervision is an important step in adaptive radiotherapy for accurate dose delivery and treatment plan re-optimization.The proposed method in this study can effectively accumulate the HDR dose to the IMRT dose domain,and the accuracy is proved to be sufficient for clinical needs.
ABSTRACT
ObjectiveTo comparison of kilo-voltage cone-beam CT (KVCBCT) deformable image registration of adaptive planning and static planning for patients with lung cancer,and evaluate their characters.MethodsFive patients with lung cancer were in the study.Two sets image were acquired every three days and were concatenated to one set.Ten sets CBCT image and planning CT image were transferred a commercial deformable image registration software.The planning CT was deformed to each set CBCT and the contours delineated,the new contour were labeled CBCTf1 -CBCTf10.Transfer of each deformed planning CT and CBCTf1 -CBCTf10 back into the treatment planning system enable re-calculation of actual dose distribution,then we obtain CT planning and fractional CBCT contour planning,the CBCT planning were labeled CBCTp1-CBCTp10.Ten times CBCT planning of every patient were added to acquire a total dose accumulation planning ( DA plan ),comparison of dose distribution and dose-volume histogram in CT plan and DA plan for fractionation dose and accumulation dose of left,right,total lung,PTV and spinal-cord.The difference of two plan was analyzed by Wilcoxson's sign rank test.ResultsThe max and min dose of PTV,the left,right,total lung V5,V10,V20,V30,V50,spinal-cord max dose,and the left,right and total lung mean dose in DA plan were smaller than in CT plan (z=-2.02 - -2.03,P＜0.05).The mean dose of PTV and D95 in DA plan was as well as in CT plan (z=-1.48,-1.21,P=0.138,0.225).ConclusionsKVCBCT based deformable image registration of adaptive planning reduce the dose of lung and spinal-cord,and enhance the dose of PTV.This provides a tool for exploring adaptive radiotherapy strategies.