Spinal Cord Contour Research Articles

Monitoring Variations in the Use of Automated Contouring Software

Our purpose was to identify variations in the clinical use of automatically generated contours that could be attributed to software error, off-label use, or automation bias. For 500 head and neck patients who were contoured by an in-house automated contouring system, Dice similarity coefficient and added path length were calculated between the contours generated by the automated system and the final contours after editing for clinical use. Statistical process control was used and control charts were generated with control limits at 3 standard deviations. Contours that exceeded the thresholds were investigated to determine the cause. Moving mean control plots were then generated to identify dosimetrists who were editing less over time, which could be indicative of automation bias. Major contouring edits were flagged for: 1.0% brain, 3.1% brain stem, 3.5% left cochlea, 2.9% right cochlea, 4.8% esophagus, 4.1% left eye, 4.0% right eye, 2.2% left lens, 4.9% right lens, 2.5% mandible, 11% left optic nerve, 6.1% right optic nerve, 3.8% left parotid, 5.9% right parotid, and 3.0% of spinal cord contours. Identified causes of editing included unexpected patient positioning, deviation from standard clinical practice, and disagreement between dosimetrist preference and automated contouring style. A statistically significant (P < .05) difference was identified between the contour editing practice of dosimetrists, with 1 dosimetrist editing more across all organs at risk. Eighteen percent (27/150) of moving mean control plots created for 5 dosimetrists indicated the amount of contour editing was decreasing over time, possibly corresponding to automation bias. The developed system was used to detect statistically significant edits caused by software error, unexpected clinical use, and automation bias. The increased ability to detect systematic errors that occur when editing automatically generated contours will improve the safety of the automatic treatment planning workflow.

Spinal Cord Delineation Based on Computed Tomography Myelogram Versus T2 Magnetic Resonance Imaging in Spinal Stereotactic Body Radiation Therapy

Spinal cord delineation is critical to the delivery of stereotactic body radiation therapy (SBRT). Although underestimating the spinal cord can lead to irreversible myelopathy, overestimating the spinal cord may compromise the planning target volume coverage. We compare spinal cord contours based on computed tomography (CT) simulation with a myelogram to spinal cord contours based on fused axial T2 magnetic resonance imaging (MRI). Eight patients with 9 spinal metastases treated with spinal SBRT were contoured by 8 radiation oncologists, neurosurgeons, and physicists, with spinal cord definition based on (1) fused axial T2 MRI and (2) CT-myelogram simulation images, yielding 72 sets of spinal cord contours. The spinal cord volume was contoured at the target vertebral body volume based on both images. The mixed-effect model assessed comparisons of T2 MRI- to myelogram-defined spinal cord in centroid deviations (deviations in the center point of the cord) through the vertebral body target volume, spinal cord volumes, and maximum doses (0.035 cc point) to the spinal cord applying the patient's SBRT treatment plan, in addition to in-between and within-subject variabilities. The estimate for the fixed effect from the mixed model showed that the mean difference between 72 CT volumes and 72 MRI volumes was 0.06 cc and was not statistically significant (95% confidence interval, -0.034, 0.153; P=.1832). The mixed model showed that the mean dose at 0.035 cc for CT-defined spinal cord contours was 1.24 Gy lower than that of MRI-defined spinal cord contours and was statistically significant (95% confidence interval, -2.292, -0.180; P=.0271). Also, the mixed model indicated no statistical significance for deviations in any of the axes between MRI-defined spinal cord contours and CT-defined spinal cord contours. CT myelogram may not be required when MRI imaging is feasible, although uncertainty at the cord-to-treatment volume interface may result in overcontouring and hence higher estimated cord dose-maximums with axial T2 MRI-based cord definition.

Study of Spinal Cord Substructure Expansion Margin in Esophageal Cancer.

Purpose:To analyze the setup errors and residual errors of different spinal cord parts in esophageal cancer patients and to explore the necessity of spinal cord segmental expansion.Methods and Materials:Sixty cases of esophageal cancer were included with 20 patients subdivided into the following groups: neck, chest and abdomen as per the treatment site. The patients underwent intensity modulated radiation therapy (IMRT) between 2017 and 2019. Thermoplastic mask or vacuum bag were utilized for immobilization of different groups. CTVision (Siemens CT-On-Rail system) was used to acquire pre-treatment CT, and 20 consecutive pre-treatment CT datasets were collected for data analysis for each case. Images were exported to MIM (MIM Software Inc.) for processing and data analysis. Dice coefficient, maximum Hausdorff distance and centroid coordinate values between the spinal cord contours in the pre-treatment CTs and the planning CT were calculated and extracted. The contour expansion margin value is calculated as MPRV = 1.3 ∑ total + 0.5 σ total, where ∑ total and σ total are the systematic and random error, respectively.Results:For neck, chest, abdominal segments of the spinal cord, the mean Dice coefficients (± SD) are 0.73 ± 0.06, 0.80 ± 0.06, 0.82 ± 0.06, the maximum Hausdorff distance residual error (± SD) are 4.46 ± 0.55, 3.49 ± 0.53, 3.46 ± 0.69 mm, and the mean centroid coordinate residual error (± SD) are 2.40 ± 0.53, 1.66 ± 0.47, 2.14 ± 0.95 mm, respectively. The calculated margin using residual centroid method in medial-lateral (ML), anterior-posterior (AP), and cranial-caudal (CC) direction of spinal cord in neck, chest, abdominal segments are 3.86, 5.37, 6.36 mm, 3.45, 3.83, 4.51 mm, 4.05, 4.83, 7.06 mm, respectively, and the calculated margin using residual Hausdorff method are 3.10, 5.33 and 6.15 mm, 3.30, 3.77, 4.61 mm, 3.35, 4.76, 6.87 mm, respectively.Conclusion:The setup errors and residual errors are different in each segment of the spinal cord. Different margins expansion should be applied to different segment of spinal cord.

Multimodality image registration in the head-and-neck using a deep learning-derived synthetic CT as a bridge.

To develop and demonstrate the efficacy of a novel head-and-neck multimodality image registration technique using deep-learning-based cross-modality synthesis. Twenty-five head-and-neck patients received magnetic resonance (MR) and computed tomography (CT) (CTaligned ) scans on the same day with the same immobilization. Fivefold cross validation was used with all of the MR-CT pairs to train a neural network to generate synthetic CTs from MR images. Twenty-four of 25 patients also had a separate CT without immobilization (CTnon-aligned ) and were used for testing. CTnon-aligned 's were deformed to the synthetic CT, and compared to CTnon-aligned registered to MR. The same registrations were performed from MR to CTnon-aligned and from synthetic CT to CTnon-aligned . All registrations used B-splines for modeling the deformation, and mutual information for the objective. Results were evaluated using the 95% Hausdorff distance among spinal cord contours, landmark error, inverse consistency, and Jacobian determinant of the estimated deformation fields. When large initial rigid misalignment is present, registering CT to MRI-derived synthetic CT aligns the cord better than a direct registration. The average landmark error decreased from 9.8±3.1mm in MR→CTnon-aligned to 6.0±2.1mm in CTsynth →CTnon-aligned deformable registrations. In the CT to MR direction, the landmark error decreased from 10.0±4.3mm in CTnon-aligned →MR deformable registrations to 6.6±2.0mm in CTnon-aligned →CTsynth deformable registrations. The Jacobian determinant had an average value of 0.98. The proposed method also demonstrated improved inverse consistency over the direct method. We showed that using a deep learning-derived synthetic CT in lieu of an MR for MR→CT and CT→MR deformable registration offers superior results to direct multimodal registration.

From Multimodality to Monomodality: Head and Neck MR-CT Registration with Synthetic Image Bridge

To develop and demonstrate the efficacy of a novel head-and-neck multimodality image registration technique using generative adversarial networks, and show its improvement over traditional direct multimodal registration. 25 head-and-neck patients received MR and CT (CTaligned) scans on the same day with the same immobilization. 19 of the MR-CT pairs were used to train a neural network to generate synthetic CTs from MR images while the registration testing was performed on the remaining 6 patients, who also had a separate CT without immobilization (CTnon-aligned). The CTnon-aligned images exhibited large neck motion. CTnon-aligned’s were deformed to the synthetic CT using b-splines and mutual information, thus effectively becoming a monomodal registration. These results were compared to CTnon-aligned directly registered to MR (multimodal). The same registrations were performed in the opposite direction. Results were evaluated using the mean distance to agreement (MDA) among spinal cord contours, landmark error, and Jacobian determinant of the estimated deformation fields. The cord contours are better matched when a synthetic CT replaces the MR, compared to a direct deformable registration with the MR. At larger initial misalignment (2-15mm MDA), the synthetic CT decreases the MDA by 1 - 10mm and 0.4 - 3mm for MR→CT and CT→MR directions, respectively. The average landmark error decreased from 10.95mm to 6.22mm when going from an MR→CTnon-aligned to a CTsynth→CTnon-aligned deformable registration. In the CT to MR direction, the average landmark error decreased from 8.19mm to 6.71mm when going from a CTnon-aligned→MR to a CTnon-aligned→CTsynth deformable registration. The proposed method demonstrated inverse consistency for both the contour and landmark analysis, while the direct method does not. Additionally, the Jacobian determinant showed feasible expansion and shrinking, with no infeasible space folding (negative values). We showed that using a synthetic CT in lieu of an MR for MR→CT and CT→MR deformable registration offers superior results to direct registration.

Lung SBRT Credentialing in the Canadian LUSTRE Randomized Trial

OCOG-LUSTRE is an ongoing phase III randomized trial (NCT01968941) comparing stereotactic body radiation therapy (SBRT) versus conventionally hypofractionated radiation for stage I medically inoperable non-small cell lung cancer (NSCLC). As part of this trial, an extensive credentialing process was undertaken due to the variability in Canadian lung SBRT practice. Here we report the final results from this initiative. A total of 15 Canadian centers participated in the SBRT credentialing process - 8 centers had previous SBRT experience (≥ 2 years) and 7 centers did not. Each center was required to satisfy three credentialing requirements: a) Completion of an institutional survey describing appropriate treatment, dosimetry, and image guidance equipment and parameters; b) Demonstration of acceptable delivery accuracy using the Imaging and Radiation Oncology Core (IROC) thoracic phantom; and c) Successful contouring and completion of standardized test cases for one central and one peripheral NSCLC. All contours and plan metrics were compared to the host institution as a nominal gold standard. All centers have successfully completed surveys and IROC phantom testing. Of the 30 SBRT test cases evaluated, 10 required a re-plan due to major deviations, mainly due to contouring errors. There were no differences in re-planning rates between established and new centers (37.5% vs 28.6%, p=0.61). Of the finally submitted plans, mean contouring errors (calculated as the distance between corresponding contours) were highest for brachial plexus in the central (C) case (12.6±6.6mm), and for the vessels in the peripheral (P) case (13.0±12.6mm). Other highly variable contours included ribs (4.9±4.9mm for C, 7.2±3.4mm for P), and proximal bronchial tree (PBT) (2.8±0.8 for C, 3.3±1.1 for P). Mean dice coefficients were similarly lowest for plexus (0.37±0.20 for C, 0.37±0.14 for P), PBT (0.77±0.06 for C, 0.75±0.09 for P), and vessels (0.69±0.29 for C, 0.64±0.31 for P), as well as esophagus (0.74±0.04 for C, 0.76±0.04 for P). Target volumes (ITV/PTV) and spinal cord contours demonstrated high concordance. All plans achieved per-protocol constraints for PTV coverage and maximum/volumetric constraints to organs-at-risk, although variations existed in dose gradients outside the target. For example, the standard deviation of the mean D2cm was 5.2 Gy (C), and 5.1 Gy (P), and the ratio of the 50% isodose volume/PTV increased by a factor of approximately 1.6 from its minimum to maximum. Our credentialing process demonstrates that small but non-negligible differences exist in both contouring and planning with lung SBRT, regardless of center experience. These variations were more likely to affect organs where concerns regarding late toxicity exist. Therefore, efforts to minimize differences through robust peer review and appropriate avenues of feedback are necessary to improve overall plan quality.

Prevalence and Imaging Characteristics of Nonmyelopathic and Myelopathic Spondylotic Cervical Cord Compression.

Cross-sectional population-based observational study. To estimate the prevalence of nonmyelopathic spondylotic cervical cord compression (NMSCCC) and cervical spondylotic myelopathy (CSM) in a population older than 40 years and to evaluate the magnetic resonance imaging (MRI) characteristics of these conditions. The prevalence of neither NMSCCC nor CSM is known and there exists no commonly accepted quantitative MRI definition of cervical cord compression. A group of 183 randomly recruited volunteers, 93 women, median age 66 years, range 40-80 years, underwent MRI examination of the cervical spine and spinal cord on a 1.5 T device using conventional sequences from disc levels C2/C3 to C6/C7. The imaging criterion for cervical cord compression was defined as a change in spinal cord contour at the level of an intervertebral disc on axial or sagittal MRI scan. MRI signs of cervical cord compression were found in 108 individuals (59.0%; 95% CI: 51.5%-66.2%); their numbers increased with age from 31.6% in the fifth decade to 66.8% in the eighth. Clinical signs of symptomatic CSM were found in two cases (1.1%), and 75 cases (41.0%) were without compression. An anteroposterior cervical canal diameter at the level of intervertebral disc (CDdisc) of less than 9.9 mm was associated with the highest probability of NMSCCC-odds ratio (OR) = 32.5, followed by a compression ratio of ≤0.5: OR = 11.1. The prevalence of NMSCCC in a population older than 40 years is higher than previously reported and increases with age. CDdisc and compression ratio had the highest capacity to discriminate between subjects with and without asymptomatic compression, and their cut-off values could be used to objectify criteria for cervical cord compression. 2.

Segmentation of the human spinal cord.

Segmenting the spinal cord contour is a necessary step for quantifying spinal cord atrophy in various diseases. Delineating gray matter (GM) and white matter (WM) is also useful for quantifying GM atrophy or for extracting multiparametric MRI metrics into specific WM tracts. Spinal cord segmentation in clinical research is not as developed as brain segmentation, however with the substantial improvement of MR sequences adapted to spinal cord MR investigations, the field of spinal cord MR segmentation has advanced greatly within the last decade. Segmentation techniques with variable accuracy and degree of complexity have been developed and reported in the literature. In this paper, we review some of the existing methods for cord and WM/GM segmentation, including intensity-based, surface-based, and image-based methods. We also provide recommendations for validating spinal cord segmentation techniques, as it is important to understand the intrinsic characteristics of the methods and to evaluate their performance and limitations. Lastly, we illustrate some applications in the healthy and pathological spinal cord. One conclusion of this review is that robust and automatic segmentation is clinically relevant, as it would allow for longitudinal and group studies free from user bias as well as reproducible multicentric studies in large populations, thereby helping to further our understanding of the spinal cord pathophysiology and to develop new criteria for early detection of subclinical evolution for prognosis prediction and for patient management. Another conclusion is that at the present time, no single method adequately segments the cord and its substructure in all the cases encountered (abnormal intensities, loss of contrast, deformation of the cord, etc.). A combination of different approaches is thus advised for future developments, along with the introduction of probabilistic shape models. Maturation of standardized frameworks, multiplatform availability, inclusion in large suite and data sharing would also ultimately benefit to the community.

Dose-escalated radiotherapy with helical tomotherapy to spine oligometastasis from intrahepatic malignancies.

325 Background: Traditional palliative radiotherapy of 30 Gy in 10 fractions to spine metastases often results in poor local control. This study evaluated the role of high dose helical tomotherapy (HT) to improve local control of spine oligometastasis from intrahepatic malignancies Methods: From 2006 to 2010, 11 hepatocellular carcinoma (HCC) patients and 3 intrahepatic cholangiocellular carcinoma patients were treated with HT for oligometastsis to spine. The gross tumor volume (GTV) was the tumor evident from MRI images fused with simulation CT images. Planning target volume (PTV) encompassed involved vertebral body or 1cm margin to GTV. The spinal cord contour represented a small area within the intradural space. To the adjacent spinal cord volume, 20 Gy dose constraint was used. Mega-voltage computed tomography (MVCT) images were used to maintain setup error within 1 mm. We assessed local control rate after HT for spine metastasis in this retrospective study. Pain response was scored by using a numerical rating scale (from 0 to 10). Results: Spine metastatic lesions were treated with median dose of 40 Gy (range 31-51 Gy) and median 5 Gy per fraction (range 3-6 Gy) to GTV and median dose of 29 Gy (range 22.5-42.5 Gy) and median 3.75 Gy per fraction( range 2.5-5 Gy) to PTV. Median biologically equivalent dose (BED, α/β=10 Gy) was 55.2 Gy (range 45.6-76.8) to GTV. Infield failure rate was 35.7% and outfield failure rate was 78.6%. Ten patients initially accompanied pain with spine metastasis (median NPIS 7.5), and all patients showed pain relief after tomotherapy. Complete relief was shown in 20% of patients and partial relief in 80 % of patients, with median NPIS reduction of 5. The patients treated with BED (α/β=10 Gy) over 56 Gy had no local failure. Median local recurrence free survival (LRFS) was 3 months. In the univariate analysis, BED (α/β=10 Gy) over 56 Gy was a prognostic factor associated with improved LRFS (p=0.008). Conclusions: HT was capable of delivering higher BED to metastatic lesions in close proximity of the spinal cord. Spine metastases from intrahepatic tumors were sensitive to high dose radiation, and BED (α/β=10Gy) higher than 56 Gy could improve local control.

SU-E-T-891: A Study of RapidArc SBRT Spine Treatment Plans

Purpose: The arc and modulating features of RapidArc treatments facilitate fast dose fall off and desirable dose distribution around the PTV and spinal cord in spine SBRT/SRS treatments. This is a study of the dosimetry characters of 11 RapidArc SBRT spine plans. Methods: All patients had treatment plans formulated on the Varian Eclipse system, and were treated on RapidArc upgraded Varian Trilogy unit, using 6MV photon beams. Patients were immobilized with the BodyFix system or the T‐frame mask system (cervical spine targets). All SBRT spine plans consisted of two arcs, each ∼350 degrees. To minimize MLC leakage and finite leaf width (0.5 cm) effects, we used collimator angles of 155 and 215 (Varian scale). Fraction doses ranged from 400 cGy to 1800 cGy. Spinal cord/cauda equina were contoured on CT myelogram or MRIimages; the spinal cord contour was expanded between 1–3 mm. Expanding the CTV by 2 mm generated the PTV, except where the PTV would intersect the spinal cord contour described above, in which case the PTV was restricted. Next, the original cord contour was again expanded by 5 mm, this time avoiding the PTV. The PTV, this second spinal cord expansion contour, as well as other OAR's were used for plan optimization. Results: This process allows for extremely rapid dose fall off. In particular, for 5 patients, the average minimum distances between 100% and 70%, and between 90% and 50% isodose lines were 2.5 mm and 3.4 mm respectively. 90% of the PTV received 100.0% of the prescription dose on average, while 1% of the PTV received 121.7% of dose. An average of 0.1 cc of spinal cord received 39.7% dose. Conclusions: We have been able to achieve improve SBRT plans for spine patients using this paradigm.

A 3T MR Imaging Investigation of the Topography of Whole Spinal Cord Atrophy in Multiple Sclerosis

Spinal cord atrophy is a common feature of MS. However, it is unknown which cord levels are most susceptible to atrophy. We performed whole cord imaging to identify the levels most susceptible to atrophy in patients with MS versus controls and also tested for differences among MS clinical phenotypes. Thirty-five patients with MS (2 with CIS, 27 with RRMS, 2 with SPMS, and 4 with PPMS phenotypes) and 27 healthy controls underwent whole cord 3T MR imaging. The spinal cord contour was segmented and assigned to bins representing each C1 to T12 vertebral level. Volumes were normalized, and group comparisons were age-adjusted. There was a trend toward decreased spinal cord volume at the upper cervical levels in PPMS/SPMS versus controls. A trend toward increased spinal cord volume throughout the cervical and thoracic cord in RRMS/CIS versus controls reached statistical significance at the T10 vertebral level. A statistically significant decrease was found in spinal cord volume at the upper cervical levels in PPMS/SPMS versus RRMS/CIS. Opposing pathologic factors impact spinal cord volume measures in MS. Patients with PPMS demonstrated a trend toward upper cervical cord atrophy. However patients with RRMS showed a trend toward increased volume at the cervical and thoracic levels, which most likely reflects inflammation or edema-related cord expansion. With the disease causing both expansion and contraction of the cord, the specificity of spinal cord volume measures for neuroprotective therapeutic effect may be limited.

The Relationships among MRI‐Defined Spinal Cord Involvement, Brain Involvement, and Disability in Multiple Sclerosis

To determine the interrelationships between MRI-defined lesion and atrophy measures of spinal cord involvement and brain involvement and their relationships to disability in a small cohort of patients with multiple sclerosis (MS). Although it is known that cervical spinal cord atrophy correlates with disability in MS, it is unknown whether it is the most important determinant when compared to other regions of the central nervous system (CNS). Furthermore, it is not clear to what extent brain and cord lesions and atrophy are related. 3T MRI of the whole brain and whole spinal cord was obtained in 21 patients with MS, including 18 with relapsing-remitting, one with secondary progressive, one with primary progressive, and one with a clinically isolated syndrome. Brain global gray and white matter volumes were segmented with Statistical Parametric Mapping 8. Spinal cord contour volume was segmented in whole by a semi-automated method with bins assigned to either the cervical or thoracic regions. All CNS volumes were normalized by the intracranial volume. Brain and cord T2 hyperintense lesions were segmented using a semi-automated edge finding tool. Among all MRI measures, only upper cervical spinal cord volume significantly correlated with Expanded Disability Status Scale score (r =-.515, P = .020). The brain cord relationships between whole or regional spinal cord volume or lesions and gray matter, white matter, or whole brain volume or whole brain lesions were generally weak and all nonsignificant. In this preliminary study of mildly disabled, treated MS patients, cervical spinal cord atrophy most strongly correlates with physical disability in MS when accounting for a wide range of other CNS measures of lesions and atrophy, including thoracic or whole spinal cord volume, and cerebral gray, white or whole brain volume. The weak relationship between spinal cord and brain lesions and atrophy may suggest that they progress rather independently in patients with MS.

SU‐GG‐J‐151: Potential Lung Dose Reduction for Minimally‐Moving Lung Lesions

Purpose: To evaluate the potential of lung sparing with IMRT for patients whose lung tumor motion range is less than 0.5 cm treated with respiratory gating at inhale. Method and Materials: A retrospective investigation on 8 patients with 4DCT scans was performed. For each patient a maximum intensity projection (MIP) data set, over 10 uniformly distributed breathing phases, was generated. The MIPs were used in contouring of GTVs, heart, and spinal cord. IMRT optimization was performed on a single phase CT data set (Average Phase or APh), representing an average mid‐ventilation phase. An average phase CT data set within 30% gating window (Average Inhale Phase or AInPh) in the proximity of the full inhale was also generated. GTVs, heart and spinal cord contours were kept the same for the APh and AInPh, while the lungs were independently contoured. The IMRT plans consist of 5‐to‐11 coplanar and/or non‐coplanar 6MV and/or 18MV beams. The dose was escalated to the maximum allowed by the organs at risk, which are spinal cord, heart and lungs. The APh optimized plan was transferred to the AInPh and dose was recomputed. Dose indices (DI) and generalized equivalent uniform doses (gEUDs) between APh and AInPh were cross‐compared. The DIs are and , and . Results: All DIs and gEUDs over the APh and AInPh for the targets, the spinal cord, and the heart are within ±1%. The DI differences in the lungs show a dose reduction ranging from 1% to over 9% for the gated plans. Conclusion: A reduction in lung dose of up to 9% is possible when gating at inhale is utilized for treatment of lung tumors with less than 0.5cm motion as seen on 4DCT. Alternatively, 9% target dose escalation shows an increase in TCP in excess of 30%.

Accurate quantification methods to evaluate cervical cord atrophy in multiple sclerosis patients

Automatic accurate measurement techniques are needed to increase reproducibility in the quantification of cervical cord area (CCA) with magnetic resonance (MR) imaging in the assessment of central nervous system (CNS) atrophy in multiple sclerosis (MS) patients. Two segmentation methods were implemented: (1) spatial mean brightness level estimation (SMBLE), and (2) partial-volume modeling (PVM). These were evaluated with the inclusion of spinal cord inclination and/or partial-volume-effect corrections. An averaged manually segmented set was considered as reference. Thirty MR studies were used to compare the different methods. A set of 15 MS patients and 15 control subjects within a two-year longitudinal study were used to evaluate cord atrophy with the best method. Statistical evaluation was made by using an intraclass correlation coefficient and Bland-Altman comparisons. Partial-volume modeling with spinal cord inclination correction and partial-volume spinal-cord contour contribution estimation was the most accurate method. The longitudinal test showed a 4% decrease in CCA in MS patients with no significant reduction in control subjects. The automatic PVM cord-segmentation approach, taking into consideration the spinal-cord inclination and partial-volume treatment, provides reproducibility and increased accuracy in the evaluation of cord atrophy, allowing the monitoring of changes in MS patients.

SU‐EE‐A3‐05: Accuracy Testing of Deformable Registration Using Dosimetric End‐Points

Purpose: To develop a dose‐based evaluation method to assess deformable image registration accuracy Method and Materials: An algorithm developed for deformable registration of MVCT to kVCT images was evaluated. The algorithm allows the generation of automatic contours on MVCT images by transferring the kVCT contours using the deformation map. The automatically generated MVCT contours can thus be used to test the deformation algorithm by comparing these contours with manual contours. Instead of a geographic contour comparison, dosimetric endpoints were evaluated after the dose distribution was calculated in the MVCT images. Three dosimetric endpoints (Dmax, Dmean, and Dose to the hottest 2 cc (Dmax (2cc)) were compared for spinal cord contours. The evaluation of geometric end‐points is directly related to the clinical information that needs to be evaluated if daily images are used for adaptive radiation therapy. A total of 93 daily megavoltage CT (MVCT) images from three patients treated for cancers in the head and neck region were evaluated. Results: Averaged over all images the calculated Dmax differed between the automatic and manual contours by 1.1 % with a standard deviation of 3.5 %. The respective values for Dmean and Dmax(2cc) are 0.1 ± 2.5 % and 1.8 ± 2.4 %. Maximum deviations between the dosimetric endpoints were 12 % for Dmax, 8% for Dmean, and 13 % for Dmax(2cc). Conclusions: Using deformable image registration, dosimetric end‐points can be generated from automatic contours in the spinal cord region that differ from manual contours by 1–2 % on average with a standard deviation of 2.5 to 3.5 %. In the spinal cord region the developed deformable image registration appears to provide sufficient accuracy to support clinical decisions. Conflict of interest: Research supported by the vendor that is commercializing the algorithm. Several co‐authors are vendor employees.

Multifractionated image-guided and stereotactic intensity-modulated radiotherapy of paraspinal tumors: A preliminary report

The use of image-guided and stereotactic intensity-modulated radiotherapy (IMRT) techniques have made the delivery of high-dose radiation to lesions within close proximity to the spinal cord feasible. This report presents clinical and physical data regarding the use of IMRT coupled with noninvasive body frames (stereotactic and image-guided) for multifractionated radiotherapy. The Memorial Sloan-Kettering Cancer Center (Memorial) stereotactic body frame (MSBF) and Memorial body cradle (MBC) have been developed as noninvasive immobilizing devices for paraspinal IMRT using stereotactic (MSBF) and image-guided (MBC) techniques. Patients were either previously irradiated or prescribed doses beyond spinal cord tolerance (54 Gy in standard fractionation) and had unresectable gross disease involving the spinal canal. The planning target volume (PTV) was the gross tumor volume with a 1 cm margin. The PTV was not allowed to include the spinal cord contour. All treatment planning was performed using software developed within the institution. Isocenter verification was performed with an in-room computed tomography scan (MSBF) or electronic portal imaging devices, or both. Patients were followed up with serial magnetic resonance imaging every 3-4 months, and no patients were lost to follow-up. Kaplan-Meier statistics were used for analysis of clinical data. Both the MSBF and MBC were able to provide setup accuracy within 2 mm. With a median follow-up of 11 months, 35 patients (14 primary and 21 secondary malignancies) underwent treatment. The median dose previously received was 3000 cGy in 10 fractions. The median dose prescribed for these patients was 2000 cGy/5 fractions (2000-3000 cGy), which provided a median PTV V100 of 88%. In previously unirradiated patients, the median prescribed dose was 7000 cGy (5940-7000 cGy) with a median PTV V100 of 90%. The median Dmax to the cord was 34% and 68% for previously irradiated and never irradiated patients, respectively. More than 90% of patients experienced palliation from pain, weakness, or paresthesia; 75% and 81% of secondary and primary lesions, respectively, exhibited local control at the time of last follow-up. No cases of radiation-induced myelopathy or radiculopathy have thus far been encountered. Precision stereotactic and image-guided paraspinal IMRT allows the delivery of high doses of radiation in multiple fractions to tumors within close proximity to the spinal cord while respecting cord tolerance. Although preliminary, the clinical results are encouraging.