Dosimetric consequences of interobserver variability in delineating the organs at risk in gynecologic interstitial brachytherapy

Generalizability of deep learning in organ-at-risk segmentation: A transfer learning study in cervical brachytherapy

PURPOSE

Deep learning can automate delineation in radiation therapy, reducing time and variability. Yet, its efficacy varies across different institutions, scanners, or settings, emphasizing the need for adaptable and robust models in clinical environments. Our study demonstrates the effectiveness of the transfer learning (TL) approach in enhancing the generalizability of deep learning models for auto-segmentation of organs-at-risk (OARs) in cervical brachytherapy.

METHODS

A pre-trained model was developed using 120 scans with ring and tandem applicator on a 3T magnetic resonance (MR) scanner (RT3). Four OARs were segmented and evaluated. Segmentation performance was evaluated by Volumetric Dice Similarity Coefficient (vDSC), 95 % Hausdorff Distance (HD95), surface DSC, and Added Path Length (APL). The model was fine-tuned on three out-of-distribution target groups. Pre- and post-TL outcomes, and influence of number of fine-tuning scans, were compared. A model trained with one group (Single) and a model trained with all four groups (Mixed) were evaluated on both seen and unseen data distributions.

RESULTS

TL enhanced segmentation accuracy across target groups, matching the pre-trained model's performance. The first five fine-tuning scans led to the most noticeable improvements, with performance plateauing with more data. TL outperformed training-from-scratch given the same training data. The Mixed model performed similarly to the Single model on RT3 scans but demonstrated superior performance on unseen data.

CONCLUSIONS

TL can improve a model's generalizability for OAR segmentation in MR-guided cervical brachytherapy, requiring less fine-tuning data and reduced training time. These results provide a foundation for developing adaptable models to accommodate clinical settings.

Safety and efficiency of a fully automatic workflow for auto-segmentation in radiotherapy using three commercially available deep learning-based applications

Advancements in radiotherapy auto-segmentation necessitate reliable and efficient workflows. Therefore, a standardized fully automatic workflow was developed for three commercially available deep learning-based auto-segmentation applications and compared to a manual workflow for safety and efficiency. The workflow underwent safety evaluation with failure mode and effects analysis. Notably, eight failure modes were reduced, including seven with severity factors ≥7, indicating the effect on patients, and two with Risk Priority Number value >125, which assesses relative risk level. Efficiency, measured by mouse clicks, showed zero clicks with the automatic workflow. This automation illustrated improvement in both safety and efficiency of workflow.

Catheters and dose optimization using a modified CVT algorithm and multi-criteria optimization in prostate HDR brachytherapy

Currently, in HDR brachytherapy planning, the catheter's positions are often selected by the planner which involves the planner's experience. The catheters are then inserted using a template which helps to guide the catheters. For certain applications, it is of interest to choose the optimal location and number of catheters needed for dose coverage and potential decrease of the treatment's toxicity. Hence, it is of great importance to develop patient-specific algorithms for catheters and dose optimization. A modified Centroidal Voronoi tessellation (CVT) algorithm is implemented and merged with a GPU-based multi-criteria optimization algorithm (gMCO). The CVT algorithm optimizes the catheters' positions, and the gMCO algorithm optimizes the dwell times and dwell positions. The CVT algorithm can be used simultaneously for insertion with or without a template. Some improvements to the CVT algorithm are presented such as a new way of considering the area that needs to be covered. One hundred and eight previously treated prostates HDR cases using real-time ultrasound (US) are used to evaluate the different optimization procedures. The plan robustness is evaluated using two types of errors; deviations (random) in the insertion and deviation (systematic) in the reconstruction of the catheters. Using gMCO on clinically inserted catheter increases the acceptance rate by 37% for RTOG criteria. Our results show that all the patients respect RTOG criteria with 11 catheters using CVT+gMCO with a template of 5 mm. The number of catheters needed for all patients to respect RTOG criteria with the freehand technique is 10 catheters using CVT+gMCO. When deviations are introduced, using a template, the acceptance rate goes to 85% with 3 mm deviations using 11 catheters. This decrease is less significant when the number of catheters is higher, decreasing by less than 5% with a 3 mm deviation using 13 catheters or more. In conclusion, it is feasible to decrease the number of catheters needed to treat most patients. Some cases still need a high number of catheters to reach the plan's criteria. Using gMCO allows an increase in the plan quality while using CVT reduces the number of catheters. A higher number of catheters equates to plans that are more robust to deviations. This article is protected by copyright. All rights reserved.

Interfraction dose deviation and catheter position in cervical interstitial and intracavitary image guided HDR brachytherapy

Interstitial and intracavitary gynecological HDR brachytherapy involve precise, localized delivery to targets with high dose gradients, sparing adjacent organs at risk (OAR). Due to the proximity of the rectum, bowel and bladder to the target, deviations in the applicator or catheter with respect to patient anatomy can significantly increase dose to OAR. The magnitude and direction of applicator and catheter migration at each fraction was assessed for template interstitial and tandem and ring (T&R) cohorts. The cohort included twelve gynecological patients with intact cervical lesions treated with external beam and brachytherapy. Pre-treatment CT images were registered to the simulation CT with respect to the target. Treatment catheter positions transformed into the planning CT coordinate system to evaluate localized catheter displacement and dose distributions calculated at each fraction. Dose was evaluated on the planning CT with planning contours and dwell locations at treatment position. Absolute deviation, depth and deflection angle for all patients were 4.6 ± 4.2 mm, -1.4 ± 4.0 mm, and 3.1 ± 2.3° respectively (n = 516 catheter positions for all treatment fractions and patients, mean ± SD). Absolute catheter deviation and deflection magnitude for interstitial treatments increased overall with each subsequent fraction with an overall increase of catheter retraction at each fraction (p < 0.005, n = 492 catheters, Kruskal-Wallis). A target EQD2 D90 reduction of 10 ± 10% and 7.7 ± 8.7% of the planned dose for interstitial and T&R cohorts respectively. There was an overall increase in bladder and rectal doses at each fraction. Catheter tracking in interstitial and intracavitary gynecological treatments with CT imaging revealed significant changes in catheter positioning with respect to the target volume. Overall deviations increased in magnitude with each subsequent fraction in the interstitial treatments. This caused patient dosimetry deviations, including target dose reduction and adjacent OAR doses changes.

Comparison of computed tomography- and magnetic resonance imaging-based target delineation for cervical cancer brachytherapy

Purpose

The objective of this study was to compare and assess the accuracy of computed tomography (CT)-based target delineation with that of magnetic resonance imaging (MRI)-based on high-dose-rate brachytherapy (HDR-BT) for patients with cervical cancer.

Material and methods

Data of 20 patients with locally advanced cervical cancer were collected and evaluated. Dimensions, conformity, and dose parameters of high-risk clinical target volume (CTV_HR) as well as D_0.1cc, D_1cc, and D_2cc of organs at risk (OARs) based on MRI were compared with those based on CT.

Results

Average age of 20 patients included was 57.8 years. Width, thickness, and volumes of CT-based CTV_HR (CTV_HR-CT) were significantly overestimated compared with those of MRI-based CTV_HR (CTV_HR-MR). Mean values of dice similarity coefficient (DSC), Hausdorff distance (HD), and centroid distance (ΔV) of CTV_HR were 0.82 cm, 0.96 cm, and 0.35 cm, respectively. Dose values of CTV_HR-CT were significantly lower compared with those of CTV_HR-MR. Concerning OARs, geometrical and dosimetric values on CT were comparable to those on MRI.

Conclusions

The delineated ranges of CTV_HR were significantly over-estimated on CT compared with MRI. D₉₈ and D₉₀ of CTV_HR-CT were lower than CTV_HR-MR. DSC and ΔV of CTV_HR and CTV_IR were similar to each other; however, there was a difference in terms of HD. CT images regarding pre-BT MR images for delineating were not enough and MRI fusion is still required.

Dose accumulation of multiple high dose rate prostate brachytherapy treatments in two commercially available image registration systems

PURPOSE

The purpose of this study was to assess whether deformable image registration (DIR) is required for dose accumulation of multiple high dose rate prostate brachytherapy (HDRPBT) plans treated with the same catheter pattern on two different CT datasets.

METHOD

DIR was applied to 20 HDRPBT patients' planning CT images who received two treatment fractions on sequential days, on two different CT datasets, with the same implant. Quality of DIR in Velocity and MIM image registration systems was assessed by calculating the Dice Similarity Coefficient (DSC) and mean distance to agreement (MDA) for the prostate, urethra and rectum contours. Accumulated doses from each system were then calculated using the same DIR technique and dose volume histogram (DVH) parameters compared to manual addition with no DIR.

RESULTS

The average DSC was found to be 0.83 (Velocity) and 0.84 (MIM), 0.80 (Velocity) and 0.80 (MIM), 0.80 (Velocity) and 0.81 (MIM), for the prostate, rectum and urethra contours, respectively. The average difference in calculated DVH parameters between the two systems using dose accumulation was less than 1%, and there was no statistically significant difference found between deformably accumulated doses in the two systems versus manual DVH addition with no DIR.

CONCLUSION

Contour propagation using DIR in velocity and MIM was shown to be at least equivalent to inter-observer contouring variability on CT. The results also indicate that dose accumulation through manual addition of DVH parameters may be sufficient for HDRPBT treatments treated with the same catheter pattern on two different CT datasets.

Defining and assessing an anisotropic delineation margin for modern radiotherapy

PURPOSE

Uncertainty in target volume delineation for modern radiotherapy impacts dosimetry and patient outcomes. Delineation uncertainty is generally overlooked in practice as a source of error, potentially since historically, other uncertainties have been the main focus. This work defined and assessed an anisotropic delineation margin in both polar and spherical coordinate systems in order to account for the spatially varying nature of this uncertainty using a whole breast radiotherapy cohort as a proof of concept.

METHODS

A cohort of 21 whole breast radiotherapy patient datasets with clinical target volumes delineated by eight independent observers was utilized. Patients were divided into categories based on target volume and laterality. An anisotropic delineation margin for each category was determined by multiplying the average standard deviation in observer contours in each category by a factor of two. Standard deviation was determined in both polar and spherical coordinates at angular increments. This anisotropic approach was compared to a conventional clinical approach, where the delineation margin was applied in the cardinal directions only. The assessment of the delineation margin was undertaken by comparing the encompassment of the observer volumes by the target volume with added margin. The extra, presumed healthy tissue included in the margin and the malignant tissue missed by the margin were determined.

RESULTS

The proposed delineation margin is effective at accounting for inter-observer variation, producing >95% coverage of all CTVs for polar, spherical, and Cartesian margins in 82%, 79%, and 92% of cases, respectively. Additionally, <1% malignant tissue was missed for 65%, 70%, and 91% of cases and <37% healthy tissue was included in 95%, 89%, and 97% of cases. A conventional delineation margin approach is most appropriate for small and gold standard target volumes. However, for large target volumes, an anisotropic margin is necessary, producing significantly greater coverage of CTVs, including significantly less presumed healthy tissue and missing significantly less malignant tissue.

CONCLUSIONS

All delineation margin methods that account for target volume and laterality proved to be adequate, with appropriate encompassment of interobserver variation and minimal inclusion of extra excess healthy tissue and exclusion of possible malignant tissue. The anisotropic approach was found to be superior to a conventional approach for target volumes >1400 cm³ only with significantly greater encompassment of interobserver variation, less missed malignant tissue and less included healthy tissue. This methodology has been validated for a whole breast radiotherapy cohort as a proof of concept, however could be applied to other anatomical sites.

A comprehensive evaluation of adaptive daily planning for cervical cancer HDR brachytherapy

The purpose of this study was to evaluate adaptive daily planning for cervical cancer patients who underwent high‐dose‐rate intracavitary brachytherapy (HDR‐BT) using comprehensive interfractional organ motion measurements. This study included 22 cervical cancer patients who underwent 5 fractions of HDR‐BT. Regions of interest (ROIs) including high‐risk clinical tumor volume (HR‐CTV) and organs at risk (OARs) were manually contoured on daily CT images. All patients were clinically treated with adaptive daily plans (ADP), which involved ROI delineation and dose optimization at each treatment fraction. Single treatment plans (SP) were retrospectively generated by applying the first treatment fraction's dwell times adjusted for decay and dwell positions of the applicator to subsequent treatment fractions. Various existing similarity metrics were calculated for the ROIs to quantify interfractional organ variations. A novel similarity (JRARM) score was established, which combined both volumetric overlap metrics (DSC, JSC, and RVD) and distance metrics (ASD, MSD, and RMSD). Linear regression was performed to determine a relationship between interfractional organ variations of various similarity metrics and D2cc variations from both plans. Wilcoxon signed‐rank tests were used to assess ADP and SP by comparing EQD2D2cc(α/β=3) for OARs. For interfractional organ variations, the sigmoid demonstrated the greatest variations based on the JRARM, DSC, and RMSD metrics. Comparisons between paired ROIs showed differences in metrics at each treatment fraction. RVD, MSD, and RMSD were found to be significantly correlated to D2cc variations for bladder and sigmoid. The comparison between plans found ADP provided lower EQD2 D2cc of OARs than SP. Specifically, the sigmoid demonstrated statistically significant dose variations (p=0.015). Substantial interfractional organ motion occurs during HDR‐BT based on comprehensive measurements and may significantly affect D2cc of OARs. Adaptive daily planning provides improved dose sparing for OARs compared to single planning with the extent of sparing being different among OARs.

PACS number(s): 87.55.D, 87.55.de, 87.55.kh, 87.57.nj

Uncertainties in volume delineation in radiation oncology: A systematic review and recommendations for future studies

BACKGROUND AND PURPOSE

Volume delineation is a well-recognised potential source of error in radiotherapy. Whilst it is important to quantify the degree of interobserver variability (IOV) in volume delineation, the resulting impact on dosimetry and clinical outcomes is a more relevant endpoint. We performed a literature review of studies evaluating IOV in target volume and organ-at-risk (OAR) delineation in order to analyse these with respect to the metrics used, reporting of dosimetric consequences, and use of statistical tests.

METHODS AND MATERIALS

Medline and Pubmed databases were queried for relevant articles using keywords. We included studies published in English between 2000 and 2014 with more than two observers.

RESULTS

119 studies were identified covering all major tumour sites. CTV (n=47) and GTV (n=38) were most commonly contoured. Median number of participants and data sets were 7 (3-50) and 9 (1-132) respectively. There was considerable heterogeneity in the use of metrics and methods of analysis. Statistical analysis of results was reported in 68% (n=81) and dosimetric consequences in 21% (n=25) of studies.

CONCLUSION

There is a lack of consistency in conducting and reporting analyses from IOV studies. We suggest a framework to use for future studies evaluating IOV.

Evaluation of an active magnetic resonance tracking system for interstitial brachytherapy

PURPOSE

In gynecologic cancers, magnetic resonance (MR) imaging is the modality of choice for visualizing tumors and their surroundings because of superior soft-tissue contrast. Real-time MR guidance of catheter placement in interstitial brachytherapy facilitates target coverage, and would be further improved by providing intraprocedural estimates of dosimetric coverage. A major obstacle to intraprocedural dosimetry is the time needed for catheter trajectory reconstruction. Herein the authors evaluate an active MR tracking (MRTR) system which provides rapid catheter tip localization and trajectory reconstruction. The authors assess the reliability and spatial accuracy of the MRTR system in comparison to standard catheter digitization using magnetic resonance imaging (MRI) and CT.

METHODS

The MRTR system includes a stylet with microcoils mounted on its shaft, which can be inserted into brachytherapy catheters and tracked by a dedicated MRTR sequence. Catheter tip localization errors of the MRTR system and their dependence on catheter locations and orientation inside the MR scanner were quantified with a water phantom. The distances between the tracked tip positions of the MRTR stylet and the predefined ground-truth tip positions were calculated for measurements performed at seven locations and with nine orientations. To evaluate catheter trajectory reconstruction, fifteen brachytherapy catheters were placed into a gel phantom with an embedded catheter fixation framework, with parallel or crossed paths. The MRTR stylet was then inserted sequentially into each catheter. During the removal of the MRTR stylet from within each catheter, a MRTR measurement was performed at 40 Hz to acquire the instantaneous stylet tip position, resulting in a series of three-dimensional (3D) positions along the catheter's trajectory. A 3D polynomial curve was fit to the tracked positions for each catheter, and equally spaced dwell points were then generated along the curve. High-resolution 3D MRI of the phantom was performed followed by catheter digitization based on the catheter's imaging artifacts. The catheter trajectory error was characterized in terms of the mean distance between corresponding dwell points in MRTR-generated catheter trajectory and MRI-based catheter digitization. The MRTR-based catheter trajectory reconstruction process was also performed on three gynecologic cancer patients, and then compared with catheter digitization based on MRI and CT.

RESULTS

The catheter tip localization error increased as the MRTR stylet moved further off-center and as the stylet's orientation deviated from the main magnetic field direction. Fifteen catheters' trajectories were reconstructed by MRTR. Compared with MRI-based digitization, the mean 3D error of MRTR-generated trajectories was 1.5 ± 0.5 mm with an in-plane error of 0.7 ± 0.2 mm and a tip error of 1.7 ± 0.5 mm. MRTR resolved ambiguity in catheter assignment due to crossed catheter paths, which is a common problem in image-based catheter digitization. In the patient studies, the MRTR-generated catheter trajectory was consistent with digitization based on both MRI and CT.

CONCLUSIONS

The MRTR system provides accurate catheter tip localization and trajectory reconstruction in the MR environment. Relative to the image-based methods, it improves the speed, safety, and reliability of the catheter trajectory reconstruction in interstitial brachytherapy. MRTR may enable in-procedural dosimetric evaluation of implant target coverage.

High-risk CTV delineation for cervix brachytherapy: Application of GEC-ESTRO guidelines in Australia and New Zealand

INTRODUCTION

Image-based brachytherapy for cervical cancer using MRI has been implemented in Australia and New Zealand. The aims of this study were to measure variability in High-risk CTV (HR-CTV) delineation and evaluate dosimetric consequences of this.

METHODS

Nine radiation oncologists, one radiation therapist and two radiologists contoured HR-CTV on 3T MRI datasets from ten consecutive patients undergoing cervical brachytherapy at a single institution. Contour comparisons were performed using the Dice Similarity Coefficient (DSC) and Mean Absolute Surface Distance (MASD). Two reference contours were created for brachytherapy planning: a Simultaneous Truth and Performance Level Estimation (STAPLE) and a consensus contour (CONSENSUS). Optimized plans (8 Gy) for both these contours were applied to individual participant's contours to assess D90 and D100 coverage of HR CTV. To compare variability in dosimetry, relative standard deviation (rSD) was calculated.

RESULTS

Good concordance (mean DSC≥0.7, MASD≤5 mm) was achieved in 8/10 cases when compared to the STAPLE reference and 6/10 cases when compared to the CONSENSUS reference. Greatest variation was visually seen in the cranio-caudal direction. The average mean rSD across all patients was 27% and 34% for the STAPLE HR-CTV D90 and D100, respectively, and 28% and 35% for the CONSENSUS HR-CTV D90 and D100. Delineation uncertainty resulted in an average dosimetric uncertainty of ±1.5-1.6 Gy per fraction based on an 8 Gy prescribed fraction.

CONCLUSIONS

Delineation of HR-CTV for cervical cancer brachytherapy was consistent amongst observers, suggesting similar interpretation of GEC-ESTRO guidelines. Despite the good concordance, there was dosimetric variation noted, which could be clinically significant.

Redesign of process map to increase efficiency: Reducing procedure time in cervical cancer brachytherapy

PURPOSE

To increase intraprocedural efficiency in the use of clinical resources and to decrease planning time for cervical cancer brachytherapy treatments through redesign of the procedure's process map.

METHODS AND MATERIALS

A multidisciplinary team identified all tasks and associated resources involved in cervical cancer brachytherapy in our institution and arranged them in a process map. A redesign of the treatment planning component of the process map was conducted with the goal of minimizing planning time. Planning time was measured on 20 consecutive insertions, of which 10 were performed with standard procedures and 10 with the redesigned process map, and results were compared. Statistical significance (p < 0.05) was measured with a two-tailed t test.

RESULTS

Twelve tasks involved in cervical cancer brachytherapy treatments were identified. The process map showed that in standard procedures, the treatment planning tasks were performed sequentially. The process map was redesigned to specify that contouring and some planning tasks are performed concomitantly. Some quality assurance tasks were reorganized to minimize adverse effects of a possible error on procedure time. Test dry runs followed by live implementation confirmed the applicability of the new process map to clinical conditions. A 29% reduction in planning time (p < 0.01) was observed with the introduction of the redesigned process map.

CONCLUSIONS

A process map for cervical cancer brachytherapy was generated. The treatment planning component of the process map was redesigned, resulting in a 29% decrease in planning time and a streamlining of the quality assurance process.