Creation of RTOG compliant patient CT-atlases for automated atlas based contouring of local regional breast and high-risk prostate cancers

Impact of a contouring atlas on radiographer inter-observer variation in male pelvis radiotherapy

PURPOSE/OBJECTIVE

To determine the impact of a MR-based contouring atlas for male pelvis radiotherapy delineation on inter-observer variation to support radiographer led real-time magnetic resonance image guided adaptive radiotherapy (MRgART).

MATERIAL/METHODS

Eight RTTs contoured 25 MR images in the Monaco treatment planning system (Monaco 5.40.01), from 5 patients. The prostate, seminal vesicles, bladder, and rectum were delineated before and after the introduction of an atlas developed through multi-disciplinary consensus. Inter-observer contour variations (volume), time to contour and observer contouring confidence were determined at both time-points using a 5-point Likert scale. Descriptive statistics were used to analyse both continuous and categorical variables. Dice similarity coefficient (DSC), Dice-Jaccard coefficient (DJC) and Hausdorff distance were used to calculate similarity between observers.

RESULTS

Although variation in volume definition decreased for all structures among all observers post intervention, the change was not statistically significant. DSC and DJC measurements remained consistent following the introduction of the atlas for all observers. The highest similarity was found in the bladder and prostate whilst the lowest was the seminal vesicles. The mean contouring time for all observers was reduced by 50% following the introduction of the atlas (53 to 27 minutes, p=0.01). For all structures across all observers, the mean contouring confidence increased significantly from 2.3 to 3.5 out of 5 (p=0.02).

CONCLUSION

Although no significant improvements were observed in contour variation amongst observers, the introduction of the consensus-based contouring atlas improved contouring confidence and speed; key factors for a real-time RTT-led MRgART.

Comparison of commercial atlas-based automatic segmentation software for prostate radiotherapy treatment planning

This study aims to assess the accuracy of automatic atlas-based contours for various key anatomical structures in prostate radiotherapy treatment planning. The evaluated structures include the bladder, rectum, prostate, seminal vesicles, femoral heads and penile bulb. CT images from 20 patients who underwent intensity-modulated radiotherapy were randomly chosen to create an atlas library. Atlas contours of the seven anatomical structures were generated using four software packages: ABAS, Eclipse, MIM, and RayStation. These contours were then compared to manual delineations performed by oncologists, which served as the ground truth. Evaluation metrics such as dice similarity coefficient (DSC), mean distance to agreement (MDA), and volume ratio (VR) were calculated to assess the accuracy of the contours. Additionally, the time taken by each software to generate the atlas contour was recorded. The mean DSC values for the bladder exhibited strong agreement (>0.8) with manual delineations for all software except for Eclipse and RayStation. Similarly, the femoral heads showed significant similarity between the atlas contours and ground truth across all software, with mean DSC values exceeding 0.9 and MDA values close to zero. On the other hand, the penile bulb displayed only moderate agreement with the ground truth, with mean DSC values ranging from 0.5 to 0.7 for all software. A similar trend was observed in the prostate atlas contours, except for MIM, which achieved a mean DSC of over 0.8. For the rectum, both ABAS and MIM atlases demonstrated strong agreement with the ground truth, resulting in mean DSC values of more than 0.8. Overall, MIM and ABAS outperformed Eclipse and RayStation in both DSC and MDA. These results indicate that the atlas-based segmentation employed in this study produces acceptable contours for the anatomical structures of interest in prostate radiotherapy treatment planning.

Development of Prostate Bed Delineation Consensus Guidelines for Magnetic Resonance Image-Guided Radiotherapy and Assessment of Its Effect on Interobserver Variability

PURPOSE

The use of magnetic resonance imaging (MRI) in radiotherapy planning is becoming more widespread, particularly with the emergence of MRI-guided radiotherapy systems. Existing guidelines for defining the prostate bed clinical target volume (CTV) show considerable heterogeneity. This study aimed to establish baseline interobserver variability (IOV) for prostate bed CTV contouring on MRI, develop international consensus guidelines, and evaluate its effect on IOV.

METHODS AND MATERIALS

Participants delineated the CTV on 3 MRI scans, obtained from the Elekta Unity MR-Linac, as per their normal practice. Radiation oncologist contours were visually examined for discrepancies, and interobserver comparisons were evaluated against simultaneous truth and performance level estimation (STAPLE) contours using overlap metrics (Dice similarity coefficient and Cohen's kappa), distance metrics (mean distance to agreement and Hausdorff distance), and volume measurements. A literature review of postradical prostatectomy local recurrence patterns was performed and presented alongside IOV results to the participants. Consensus guidelines were collectively constructed, and IOV assessment was repeated using these guidelines.

RESULTS

Sixteen radiation oncologists' contours were included in the final analysis. Visual evaluation demonstrated significant differences in the superior, inferior, and anterior borders. Baseline IOV assessment indicated moderate agreement for the overlap metrics while volume and distance metrics demonstrated greater variability. Consensus for optimal prostate bed CTV boundaries was established during a virtual meeting. After guideline development, a decrease in IOV was observed. The maximum volume ratio decreased from 4.7 to 3.1 and volume coefficient of variation reduced from 40% to 34%. The mean Dice similarity coefficient rose from 0.72 to 0.75 and the mean distance to agreement decreased from 3.63 to 2.95 mm.

CONCLUSIONS

Interobserver variability in prostate bed contouring exists among international genitourinary experts, although this is lower than previously reported. Consensus guidelines for MRI-based prostate bed contouring have been developed, and this has resulted in an improvement in contouring concordance. However, IOV persists and strategies such as an education program, development of a contouring atlas, and further refinement of the guidelines may lead to additional improvements.

Extensive clinical testing of Deep Learning Segmentation models for thorax and breast cancer radiotherapy planning

BACKGROUND

The performance of deep learning segmentation (DLS) models for automatic organ extraction from CT images in the thorax and breast regions was investigated. Furthermore, the readiness and feasibility of integrating DLS into clinical practice were addressed by measuring the potential time savings and dosimetric impact.

MATERIAL AND METHODS

Thirty patients referred to radiotherapy for breast cancer were prospectively included. A total of 23 clinically relevant left- and right-sided organs were contoured manually on CT images according to ESTRO guidelines. Next, auto-segmentation was executed, and the geometric agreement between the auto-segmented and manually contoured organs was qualitatively assessed applying a scale in the range [0-not acceptable, 3-no corrections]. A quantitative validation was carried out by calculating Dice coefficients (DSC) and the 95% percentile of Hausdorff distances (HD95). The dosimetric impact of optimizing the treatment plans on the uncorrected DLS contours, was investigated from a dose coverage analysis using DVH values of the manually delineated contours as references.

RESULTS

The qualitative analysis showed that 93% of the DLS generated OAR contours did not need corrections, except for the heart where 67% of the contours needed corrections. The majority of DLS generated CTVs needed corrections, whereas a minority were deemed not acceptable. Still, using the DLS-model for CTV and heart delineation is on average 14 minutes faster. An average DSC=0.91 and H95=9.8 mm were found for the left and right breasts, respectively. Likewise, and average DSC in the range [0.66, 0.76]mm and HD95 in the range [7.04, 12.05]mm were found for the lymph nodes.

CONCLUSION

The validation showed that the DLS generated OAR contours can be used clinically. Corrections were required to most of the DLS generated CTVs, and therefore warrants more attention before possibly implementing the DLS models clinically.

Automated Segmentation of the Clinical Target Volume in the Planning CT for Breast Cancer Using Deep Neural Networks

3-D radiotherapy is an effective treatment modality for breast cancer. In 3-D radiotherapy, delineation of the clinical target volume (CTV) is an essential step in the establishment of treatment plans. However, manual delineation is subjective and time consuming. In this study, we propose an automated segmentation model based on deep neural networks for the breast cancer CTV in planning computed tomography (CT). Our model is composed of three stages that work in a cascade manner, making it applicable to real-world scenarios. The first stage determines which slices contain CTVs, as not all CT slices include breast lesions. The second stage detects the region of the human body in an entire CT slice, eliminating boundary areas, which may have side effects for the segmentation of the CTV. The third stage delineates the CTV. To permit the network to focus on the breast mass in the slice, a novel dynamically strided convolution operation, which shows better performance than standard convolution, is proposed. To train and evaluate the model, a large dataset containing 455 cases and 50 425 CT slices is constructed. The proposed model achieves an average dice similarity coefficient (DSC) of 0.802 and 0.801 for right-0 and left-sided breast, respectively. Our method shows superior performance to that of previous state-of-the-art approaches.

Clinical validation of an automatic atlas‐based segmentation tool for male pelvis CT images

Purpose

This retrospective work aims to evaluate the possible impact on intra‐ and inter‐observer variability, contouring time, and contour accuracy of introducing a pelvis computed tomography (CT) auto‐segmentation tool in radiotherapy planning workflow.

Methods

Tests were carried out on five structures (bladder, rectum, pelvic lymph‐nodes, and femoral heads) of six previously treated subjects, enrolling five radiation oncologists (ROs) to manually re‐contour and edit auto‐contours generated with a male pelvis CT atlas created with the commercial software MIM MAESTRO. The ROs first delineated manual contours (M). Then they modified the auto‐contours, producing automatic‐modified (AM) contours. The procedure was repeated to evaluate intra‐observer variability, producing M1, M2, AM1, and AM2 contour sets (each comprising 5 structures × 6 test patients × 5 ROs = 150 contours), for a total of 600 contours. Potential time savings was evaluated by comparing contouring and editing times. Structure contours were compared to a reference standard by means of Dice similarity coefficient (DSC) and mean distance to agreement (MDA), to assess intra‐ and inter‐observer variability. To exclude any automation bias, ROs evaluated both M and AM sets as “clinically acceptable” or “to be corrected” in a blind test.

Results

Comparing AM to M sets, a significant reduction of both inter‐observer variability (p < 0.001) and contouring time (‐45% whole pelvis, p < 0.001) was obtained. Intra‐observer variability reduction was significant only for bladder and femoral heads (p < 0.001). The statistical test showed no significant bias.

Conclusion

Our atlas‐based workflow proved to be effective for clinical practice as it can improve contour reproducibility and generate time savings. Based on these findings, institutions are encouraged to implement their auto‐segmentation method.

The dosimetric impact of axillary nodes contouring variability in breast cancer radiotherapy: an AIRO multi-institutional study

AIM

To quantify the dosimetric impact of contouring variability of axillary lymph nodes (L2, L3, L4) in breast cancer (BC) locoregional radiotherapy (RT).

MATERIALS AND METHODS

18 RT centres were asked to plan a locoregional treatment on their own planning target volume (single centre, SC-PTV) which was created by applying their institutional margins to the clinical target volume of the axillary nodes of three BC patients (P1, P2, P3) previously delineated (SC-CTV). The gold standard CTVs (GS-CTVs) of P1, P2 and P3 were developed by BC experts' consensus and validated with STAPLE algorithm. For each participating centre, the GS-PTV of each patient was created by applying the same margins as those used for the SC-CTV to SC-PTV expansion and replaced the SC-PTV in the treatment plan. Datasets were imported into MIM v6.1.7 [MIM Software Inc.], where dose-volume histograms (DVHs) were extracted and differences were analysed.

RESULTS

17/18 centres used intensity-modulated RT (IMRT). The CTV to PTV margins ranged from 0 to 10 mm (median 5 mm). No correlation was observed between GS-CTV coverage by 95% isodose and GS-PTV margins width. Doses delivered to 98% (D98) and 95% (D95) of GS-CTVs were significantly lower than those delivered to the SC-CTVs. No significant difference between SC-CTV and GS-CTV was observed in maximum dose (D2), always under 110%. Mean dose ≥ 99% of the SC-CTVs and GS-CTVs was satisfied in 84% and 50%, respectively. In less than one half of plans, GS-CTV V95% was above 90%. Breaking down the GS-CTV into the three nodal levels (L2, L3 and L4), L4 had the lowest probability to be covered by the 95% isodose.

CONCLUSIONS

Overall, GS-CTV resulted worse coverage, especially for L4. IMRT was largely used and CTV-to-PTV margins did not compensate for contouring issues. The results highlighted the need for delineation training and standardization.

A comparative study of auto-contouring softwares in delineation of organs at risk in lung cancer and rectal cancer

Radiotherapy requires the target area and the organs at risk to be contoured on the CT image of the patient. During the process of organs-at-Risk (OAR) of the chest and abdomen, the doctor needs to contour at each CT image. The delineations of large and varied shapes are time-consuming and laborious. This study aims to evaluate the results of two automatic contouring softwares on OARs definition of CT images of lung cancer and rectal cancer patients. The CT images of 15 patients with rectal cancer and 15 patients with lung cancer were selected separately, and the organs at risk were manually contoured by experienced physicians as reference structures. And then the same datasets were automatically contoured based on AiContour (version 3.1.8.0, Manufactured by Linking MED, Beijing, China) and Raystation (version 4.7.5.4, Manufactured by Raysearch, Stockholm, Sweden) respectively. Deep learning auto-segmentations and Atlas were respectively performed with AiContour and Raystation. Overlap index (OI), Dice similarity index (DSC) and Volume difference (Dv) were evaluated based on the auto-contours, and independent-sample t-test analysis is applied to the results. The results of deep learning auto-segmentations on OI and DSC were better than that of Atlas with statistical difference. There was no significant difference in Dv between the results of two software. With deep learning auto-segmentations, auto-contouring results of most organs in the chest and abdomen are good, and with slight modification, it can meet the clinical requirements for planning. With Atlas, auto-contouring results in most OAR is not as good as deep learning auto-segmentations, and only the auto-contouring results of some organs can be used clinically after modification.

A novel specific grading standard study of auto-segmentation of organs at risk in thorax: subjective-objective-combined grading standard

BACKGROUND

To develop a novel subjective-objective-combined (SOC) grading standard for auto-segmentation for each organ at risk (OAR) in the thorax.

METHODS

A radiation oncologist manually delineated 13 thoracic OARs from computed tomography (CT) images of 40 patients. OAR auto-segmentation accuracy was graded by five geometric objective indexes, including the Dice similarity coefficient (DSC), the difference of the Euclidean distance between centers of mass (ΔCMD), the difference of volume (ΔV), maximum Hausdorff distance (MHD), and average Hausdorff distance (AHD). The grading results were compared with those of the corresponding geometric indexes obtained by geometric objective methods in the other two centers. OAR auto-segmentation accuracy was also graded by our subjective evaluation standard. These grading results were compared with those of DSC. Based on the subjective evaluation standard and the five geometric indexes, the correspondence between the subjective evaluation level and the geometric index range was established for each OAR.

RESULTS

For ΔCMD, ΔV, and MHD, the grading results of the geometric objective evaluation methods at our center and the other two centers were inconsistent. For DSC and AHD, the grading results of three centers were consistent. Seven OARs' grading results in the subjective evaluation standard were inconsistent with those of DSC. Six OARs' grading results in the subjective evaluation standard were consistent with those of DSC. Finally, we proposed a new evaluation method that combined the subjective evaluation level of those OARs with the range of corresponding DSC to determine the grading standard. If the DSC ranges between the adjacent levels did not overlap, the DSC range was used as the grading standard. Otherwise, the mean value of DSC was used as the grading standard.

CONCLUSIONS

A novel OAR-specific SOC grading standard in thorax was developed. The SOC grading standard provides a possible alternative for evaluation of the auto-segmentation accuracy for thoracic OARs.

Visualizing the urethra by magnetic resonance imaging without usage of a catheter for radiotherapy of prostate cancer

•

Post urination MRI is useful for urethra-sparing radiotherapy treatment planning.

•

This prospective clinical trial included 11 prostate cancer patients.

•

Post urination MRI is the identification method of prostatic urinary tract in non-invasive manner.

The urethra position may shift due to the presence/absence of the catheter. Our proposed post-urination-magnetic resonance imaging (PU-MRI) technique is possible to identify the urethra without catheter. We aimed to verify the inter-operator difference in contouring the urethra by PU-MRI. The mean values of the evaluation indices of dice similarity coefficient, mean slice-wise Hausdorff distance, and center coordinates were 0.93, 0.17 mm, and 0.36 mm for computed tomography, and 0.75, 0.44 mm, and 1.00 mm for PU-MRI. Therefore, PU-MRI might be useful for identifying the prostatic urinary tract without using a urethral catheter.

Clinical trial registration: Hokkaido University Hospital for Clinical Research (018-0221).

Methodological approach to create an atlas using a commercial auto-contouring software

PURPOSE

The aim of this work was to establish a methodological approach for creation and optimization of an atlas for auto-contouring, using the commercial software MIM MAESTRO (MIM Software Inc. Cleveland OH).

METHODS

A computed tomography (CT) male pelvis atlas was created and optimized to evaluate how different tools and options impact on the accuracy of automatic segmentation. Pelvic lymph nodes (PLN), rectum, bladder, and femurs of 55 subjects were reviewed for consistency by a senior consultant radiation oncologist with 15 yr of experience. Several atlas and workflow options were tuned to optimize the accuracy of auto-contours. The deformable image registration (DIR), the finalization method, the k number of atlas best matching subjects, and several post-processing options were studied. To test our atlas performances, automatic and reference manual contours of 20 test subjects were statistically compared based on dice similarity coefficient (DSC) and mean distance to agreement (MDA) indices. The effect of field of view (FOV) reduction on auto-contouring time was also investigated.

RESULTS

With the optimized atlas and workflow, DSC and MDA median values of bladder, rectum, PLN, and femurs were 0.91 and 1.6 mm, 0.85 and 1.6 mm, 0.85 and 1.8 mm, and 0.96 and 0.5 mm, respectively. Auto-contouring time was more than halved by strictly cropping the FOV of the subject to be contoured to the pelvic region.

CONCLUSION

A statistically significant improvement of auto-contours accuracy was obtained using our atlas and optimized workflow instead of the MIM Software pelvic atlas.

Consensus on Contouring Primary Breast Tumors on MRI in the Setting of Neoadjuvant Partial Breast Irradiation in Trials

PURPOSE

Our purpose was to present and evaluate expert consensus on contouring primary breast tumors on magnetic resonance imaging (MRI) in the setting of neoadjuvant partial breast irradiation in trials.

METHODS AND MATERIALS

Expert consensus on contouring guidelines for target definition of primary breast tumors on contrast-enhanced MRI in trials was developed by an international team of experienced breast radiation oncologists and a dedicated breast radiologist during 3 meetings. At the first meeting, draft guidelines were developed through discussing and contouring 2 cases. At the second meeting 6 breast radiation oncologists delineated gross tumor volume (GTV) in 10 patients with early-stage breast cancer (cT1N0) according to draft guidelines. GTV was expanded isotropically (20 mm) to generate clinical target volume (CTV), excluding skin and chest wall. Delineations were reviewed for disagreement and guidelines were clarified accordingly. At the third meeting 5 radiation oncologists redelineated 6 cases using consensus-based guidelines. Interobserver variation of GTV and CTV was assessed using generalized conformity index (CI). CI was calculated as the sum of volumes each pair of observers agreed upon, divided by the sum of encompassing volumes for each pair of observers.

RESULTS

For the 2 delineation sessions combined, mean GTV ranged between 0.19 and 2.44 cm3, CI for GTV ranged between 0.28 and 0.77, and CI for CTV between 0.77 and 0.94. The largest interobserver variation in GTV delineations was observed in cases with extended tumor spiculae, blood vessels near or markers within the tumor, or with increased enhancement of glandular breast tissue. Consensus-based guidelines stated to delineate all visible tumors on contrast enhanced-MRI scan 1 to 2 minutes after contrast injection and if a marker was inserted in the tumor to include this.

CONCLUSIONS

Expert-based consensus on contouring primary breast tumors on MRI in trials has been reached. This resulted in low interobserver variation for CTV in the context of a uniform 20 mm GTV to CTV expansion margin.

An experimental and algorithm-based study of the spectral features of breast cancer patients by a photodiagnosis approach

In the present study, the spectral diagnosis of blood plasma samples of breast cancer patients and an equal number of normal controls was investigated. A set of ratio parameters was acquired by employing SXS and FES. The samples were also analyzed statistically by employing Welch two-sample t-tests, and the effects of three ratio parameters, R₁, R₂, and R₃, were also studied by plotting them against the subject numbers. A linear discriminant was also applied to verify the exact classification of normal control and breast cancer patients. It was observed that the levels of biofluorophores such as porphyrin, NADH, tryptophan and flavins were elevated 2- to 3-fold for breast cancer patients compared to normal controls, with an accuracy of approximately 100 %. We have also confirmed the validity of the obtained experimental results by using an advanced robust diagnostic algorithm. The experimental results of the current study may have a vital and substantial impact on the detection and screening protocols used for future breast cancer patients. The spectral analysis of body fluid could be of great value to add to and enhance the current procedures with an accuracy of approximately 100 % with limited number of samples. The results and objectives of this preliminary study were encouraging and useful for the discrimination of the features of breast cancer patients compared to those of normal controls.

Atlas-based auto-segmentation for postoperative radiotherapy planning in endometrial and cervical cancers

Background

Since intensity-modulated radiation therapy (IMRT) has become popular for the treatment of gynecologic cancers, the contouring process has become more critical. This study evaluated the feasibility of atlas-based auto-segmentation (ABAS) for contouring in patients with endometrial and cervical cancers.

Methods

A total of 75 sets of planning CT images from 75 patients were collected. Contours for the pelvic nodal clinical target volume (CTV), femur, and bladder were carefully generated by two skilled radiation oncologists. Of 75 patients, 60 were randomly registered in three different atlas libraries for ABAS in groups of 20, 40, or 60. ABAS was conducted in 15 patients, followed by manual correction (ABAS_c). The time required to generate all contours was recorded, and the accuracy of segmentation was assessed using Dice’s coefficient (DC) and the Hausdorff distance (HD) and compared to those of manually delineated contours.

Results

For ABAS-CTV, the best results were achieved with groups of 60 patients (DC, 0.79; HD, 19.7 mm) and the worst results with groups of 20 patients (DC, 0.75; p = 0.012; HD, 21.3 mm; p = 0.002). ABAS_c-CTV performed better than ABAS-CTV in terms of both HD and DC (ABAS_c [n = 60]; DC, 0.84; HD, 15.6 mm; all p < 0.017). ABAS required an average of 45.1 s, whereas ABAS_c required 191.1 s; both methods required less time than the manual methods (p < 0.001). Both ABAS-Femur and simultaneous ABAS-Bilateral-femurs showed satisfactory performance, regardless of the atlas library used (DC > 0.9 and HD ≤10.0 mm), with significant time reduction compared to that needed for manual delineation (p < 0.001). However, ABAS-Bladder did not prove to be feasible, with inferior results regardless of library size (DC < 0.6 and HD > 40 mm). Furthermore, ABAS_c-Bladder required a longer processing time than manual contouring to achieve the same accuracy.

Conclusions

ABAS could help physicians to delineate the CTV and organs-at-risk (e.g., femurs) in IMRT planning considering its consistency, efficacy, and accuracy.

Using infrared depth-sensing technology to improve the brachytherapy operating room experience

PURPOSE

The purpose of this study was to discuss the merits of using depth-sensing infrared camera technology in the brachytherapy operating room during interstitial brachytherapy for gynecologic malignancies.

MATERIALS AND METHODS

The infrared depth-sensing camera from a Microsoft Kinect that had been adapted for surgical use was introduced into a high-volume interstitial brachytherapy operating room. Brachytherapists then used the touchless, gestural interface to review preoperative MRI in real time to guide needle insertion.

RESULTS

The interface was used for 10 consecutive procedures by 4 separate brachytherapists. The initial training and adjustment to the technology was variable among brachytherapists. All brachytherapists found the controls intuitive and were able to successfully navigate MRI on the system after 1, 30, 30, and 45 min. Qualitatively, brachytherapists found the system helpful for interpretation of intraoperative ultrasound imaging. Furthermore, it ensured adequate needle positioning and deposition was maintained for large tumors. Surgeons involved in its use agreed on potential for considerable benefit when performing interstitial brachytherapy.

CONCLUSIONS

Adapting this technology for use in the brachytherapy suite provided a higher level of comfort with interstitial catheter placement. This novel tool or similar technology might be considered within other brachytherapy suites.

Automated planning of whole breast irradiation using hybrid IMRT improves efficiency and quality

Purpose

To develop an automated workflow for whole breast irradiation treatment planning using hybrid intensity modulated radiation therapy (IMRT) approach and to demonstrate that this workflow can improve planning quality and efficiency when compared to manual planning.

Methods

The auto planning framework was built based on scripting with MIM and Pinnacle systems. MIM workflows were developed to automatically segment normal structures and targets, identify landmarks for beam placement, select beam energies, and set beam configurations. Pinnacle scripts were generated from the MIM workflow to create hybrid IMRT plans automatically. Each hybrid IMRT plan included two prescriptions: a three‐dimensional (3D) prescription consisted of two open tangent beams, and an IMRT prescription consisted of two step‐and‐shoot IMRT beams. The 3D prescription delivered a full prescription dose to the maximum dose point, and the IMRT prescription was optimized to deliver a uniform dose to the entire breast while sparing dose to the normal structures. For 30 patients, the auto plans were compared with clinically accepted manual plans using the paired sample t‐test.

Results

The auto planning process took approximately 8 min to complete. The mean dice coefficients between auto‐segmentation and manual contours were 0.98, 0.94 and 0.88 for the lungs, heart, and PTVeval_Breast, respectively. The MUs of the auto plans was on average 13% higher than that of the manual plans. Auto planning improved plan quality significantly: percentage volume receiving 95% of the prescription dose (V95%) of the PTVeval_Breast increased from 91.5% to 93.2% (P = 0.001), V105% of the PTVeval_Breast decreased from 7.2% to 1.2% (P = 0.013), V20Gy of the ipsilateral lung decreased from 13.1% to 10.4% (P = 0.001) and mean heart dose for left‐sided breast patients decreased from 1.2 Gy to 0.9 Gy (P < 0.001).

Conclusion

An automated treatment planning process can make the planning process efficient with improved plan quality.

AAR-RT - A system for auto-contouring organs at risk on CT images for radiation therapy planning: Principles, design, and large-scale evaluation on head-and-neck and thoracic cancer cases

Contouring (segmentation) of Organs at Risk (OARs) in medical images is required for accurate radiation therapy (RT) planning. In current clinical practice, OAR contouring is performed with low levels of automation. Although several approaches have been proposed in the literature for improving automation, it is difficult to gain an understanding of how well these methods would perform in a realistic clinical setting. This is chiefly due to three key factors - small number of patient studies used for evaluation, lack of performance evaluation as a function of input image quality, and lack of precise anatomic definitions of OARs. In this paper, extending our previous body-wide Automatic Anatomy Recognition (AAR) framework to RT planning of OARs in the head and neck (H&N) and thoracic body regions, we present a methodology called AAR-RT to overcome some of these hurdles. AAR-RT follows AAR's 3-stage paradigm of model-building, object-recognition, and object-delineation. Model-building: Three key advances were made over AAR. (i) AAR-RT (like AAR) starts off with a computationally precise definition of the two body regions and all of their OARs. Ground truth delineations of OARs are then generated following these definitions strictly. We retrospectively gathered patient data sets and the associated contour data sets that have been created previously in routine clinical RT planning from our Radiation Oncology department and mended the contours to conform to these definitions. We then derived an Object Quality Score (OQS) for each OAR sample and an Image Quality Score (IQS) for each study, both on a 1-to-10 scale, based on quality grades assigned to each OAR sample following 9 key quality criteria. Only studies with high IQS and high OQS for all of their OARs were selected for model building. IQS and OQS were employed for evaluating AAR-RT's performance as a function of image/object quality. (ii) In place of the previous hand-crafted hierarchy for organizing OARs in AAR, we devised a method to find an optimal hierarchy for each body region. Optimality was based on minimizing object recognition error. (iii) In addition to the parent-to-child relationship encoded in the hierarchy in previous AAR, we developed a directed probability graph technique to further improve recognition accuracy by learning and encoding in the model "steady" relationships that may exist among OAR boundaries in the three orthogonal planes. Object-recognition: The two key improvements over the previous approach are (i) use of the optimal hierarchy for actual recognition of OARs in a given image, and (ii) refined recognition by making use of the trained probability graph. Object-delineation: We use a kNN classifier confined to the fuzzy object mask localized by the recognition step and then fit optimally the fuzzy mask to the kNN-derived voxel cluster to bring back shape constraint on the object. We evaluated AAR-RT on 205 thoracic and 298 H&N (total 503) studies, involving both planning and re-planning scans and a total of 21 organs (9 - thorax, 12 - H&N). The studies were gathered from two patient age groups for each gender - 40-59 years and 60-79 years. The number of 3D OAR samples analyzed from the two body regions was 4301. IQS and OQS tended to cluster at the two ends of the score scale. Accordingly, we considered two quality groups for each gender - good and poor. Good quality data sets typically had OQS ≥ 6 and had distortions, artifacts, pathology etc. in not more than 3 slices through the object. The number of model-worthy data sets used for training were 38 for thorax and 36 for H&N, and the remaining 479 studies were used for testing AAR-RT. Accordingly, we created 4 anatomy models, one each for: Thorax male (20 model-worthy data sets), Thorax female (18 model-worthy data sets), H&N male (20 model-worthy data sets), and H&N female (16 model-worthy data sets). On "good" cases, AAR-RT's recognition accuracy was within 2 voxels and delineation boundary distance was within ∼1 voxel. This was similar to the variability observed between two dosimetrists in manually contouring 5-6 OARs in each of 169 studies. On "poor" cases, AAR-RT's errors hovered around 5 voxels for recognition and 2 voxels for boundary distance. The performance was similar on planning and replanning cases, and there was no gender difference in performance. AAR-RT's recognition operation is much more robust than delineation. Understanding object and image quality and how they influence performance is crucial for devising effective object recognition and delineation algorithms. OQS seems to be more important than IQS in determining accuracy. Streak artifacts arising from dental implants and fillings and beam hardening from bone pose the greatest challenge to auto-contouring methods.

Evaluation of a mobile C-arm cone-beam CT in interstitial high-dose-rate prostate brachytherapy treatment planning

Introduction

The aim of this study was to evaluate the suitability of using cone‐beam computed tomography (CBCT) obtained with a mobile C‐arm X‐ray fluoroscopy unit as a single modality for planning of high‐dose‐rate (HDR) prostate brachytherapy treatments.

Methods

The feasibility of using CBCT images obtained using a Siemens Arcadis Orbic 3D mobile C‐arm was evaluated. A retrospective clinical study was undertaken of six participants undergoing HDR prostate brachytherapy. Plans generated using images from a Toshiba Aquilion One LB CT were compared with those generated using CBCT images. After rigid spatial registration, the plans were compared based on various parameters such as dose‐volume histograms, overlap quantities and metrics, and dose constraints.

Results

Provided they were within the limited field of view, the brachytherapy catheters and fiducial markers were clearly visible in the CBCT images and thus, localisable and identifiable in the treatment planning process. The Siemens CBCT underestimated CT numbers leading to poorer tissue contrast which exacerbated the difficulties in delineation of the target tumour and the surrounding organs at risk. Between CT‐ and CBCT‐based plans, the mean difference of CTV‐D₉₀ was 1.58 Gy, CTV‐V₁₀₀ was 12.13%, rectum‐V₈₀ was 0.06% and urethra‐V₁₂₀ was −0.70%.

Conclusion

It was not feasible to solely utilise the Siemens Arcadis Orbic 3D for HDR prostate brachytherapy treatment planning due to suboptimal organ delineation. However, the methods in this study could be used to evaluate other mobile CBCT imaging devices for feasibility in HDR brachytherapy treatment planning since the results indicated that it may not be necessary to have standard quality CT images for treatment planning.

Fully automatic and robust segmentation of the clinical target volume for radiotherapy of breast cancer using big data and deep learning

PURPOSE

To train and evaluate a very deep dilated residual network (DD-ResNet) for fast and consistent auto-segmentation of the clinical target volume (CTV) for breast cancer (BC) radiotherapy with big data.

METHODS

DD-ResNet was an end-to-end model enabling fast training and testing. We used big data comprising 800 patients who underwent breast-conserving therapy for evaluation. The CTV were validated by experienced radiation oncologists. We performed a fivefold cross-validation to test the performance of the model. The segmentation accuracy was quantified by the Dice similarity coefficient (DSC) and the Hausdorff distance (HD). The performance of the proposed model was evaluated against two different deep learning models: deep dilated convolutional neural network (DDCNN) and deep deconvolutional neural network (DDNN).

RESULTS

Mean DSC values of DD-ResNet (0.91 and 0.91) were higher than the other two networks (DDCNN: 0.85 and 0.85; DDNN: 0.88 and 0.87) for both right-sided and left-sided BC. It also has smaller mean HD values of 10.5 mm and 10.7 mm compared with DDCNN (15.1 mm and 15.6 mm) and DDNN (13.5 mm and 14.1 mm). Mean segmentation time was 4 s, 21 s and 15 s per patient with DDCNN, DDNN and DD-ResNet, respectively. The DD-ResNet was also superior with regard to results in the literature.

CONCLUSIONS

The proposed method could segment the CTV accurately with acceptable time consumption. It was invariant to the body size and shape of patients and could improve the consistency of target delineation and streamline radiotherapy workflows.

Fusion of Intraoperative Transrectal Ultrasound Images with Post-implant Computed Tomography and Magnetic Resonance Imaging

Purpose

To compare the impact of the fusion of intraoperative transrectal ultrasound (TRUS) images with day 30 computed tomography (CT) and magnetic resonance imaging (MRI) on prostate volume and dosimetry.

Methods and materials

Seventy-five consecutive patients with CT and MRI obtained on day 30 with a Fast Spin Echo T2-weighted magnetic resonance (MR) sequence were analyzed. A rigid manual registration was performed between the intraoperative TRUS and day-30 CT based on the prostate volume. A second manual rigid registration was performed between the intraoperative TRUS and the day-30 MRI. The prostate contours were manually modified on CT and MRI. The difference in prostate volume and dosimetry between CT and MRI were compared.

Results

Prostate volume was on average 8% (standard deviation (SD) ± 16%) larger on intraoperative TRUS than on CT and 6% (18%) larger than on MRI. In 48% of the cases, the difference in volume on CT was > 10% compared to MRI. The difference in prostate volume between CT and MRI was inversely correlated to the difference in D90 (minimum dose that covers 90% of the prostate volume) between CT and MRI (r = -0.58, P < .001). A D90 < 90% was found in 5% (n = 4) on MRI and in 10% (n = 7) on CT (Fisher exact test one-sided P = .59), but in no patient was the D90 < 90% on both MRI and CT.

Conclusions

When fusing TRUS images with CT and MRI, the differences in prostate volume between those modalities remain clinically important in nearly half of the patients, and this has a direct influence on how implant quality is evaluated.

Magnetic Resonance Imaging-Guided Adaptive Radiation Therapy: A "Game Changer" for Prostate Treatment?

Radiation therapy to the prostate involves increasingly sophisticated delivery techniques and changing fractionation schedules. With a low estimated α/β ratio, a larger dose per fraction would be beneficial, with moderate fractionation schedules rapidly becoming a standard of care. The integration of a magnetic resonance imaging (MRI) scanner and linear accelerator allows for accurate soft tissue tracking with the capacity to replan for the anatomy of the day. Extreme hypofractionation schedules become a possibility using the potentially automated steps of autosegmentation, MRI-only workflow, and real-time adaptive planning. The present report reviews the steps involved in hypofractionated adaptive MRI-guided prostate radiation therapy and addresses the challenges for implementation.

Variability in axillary lymph node delineation for breast cancer radiotherapy in presence of guidelines on a multi-institutional platform

AIM

To quantify the variability between radiation oncologists (ROs) when outlining axillary nodes in breast cancer.

MATERIAL AND METHODS

For each participating center, three ROs with different levels of expertise, i.e., junior (J), senior (S) and expert (E), contoured axillary nodal levels (L1, L2, L3 and L4) on the CT images of three different patients (P) of an increasing degree of anatomical complexity (from P1 to P2 to P3), according to contouring guidelines. Consensus contours were generated using the simultaneous truth and performance level estimation (STAPLE) method.

RESULTS

Fifteen centers and 42 ROs participated. Overall, the median Dice similarity coefficient was 0.66. Statistically significant differences were observed according to the level of expertise (better agreement for J and E, worse for S); the axillary level (better agreement for L1 and L4, worse for L3); the patient (better agreement for P1, worse for P3). Statistically significant differences in contouring were found in 18% of the inter-center comparison. Less than a half of the centers could claim to have a good agreement between the internal ROs.

CONCLUSIONS

The overall intra-institute and inter-institute agreement was moderate. Central lymph-node levels were the most critical and variability increased as the complexity of the patient's anatomy increased. These findings might have an effect on the interpretation of results from multicenter and even mono-institute studies.

Prostate cancer: contouring target and organs at risk by kilovoltage and megavoltage CT and MRI in patients with and without hip prostheses

OBJECTIVE

In radiotherapy treatment, planning target volume and organs at risk are contoured on kilovoltage CT (kVCT) images. Unlike MR images, kVCT does not provide precise information on target volume extension. Since neither kVCT nor MRI may be suitable for contouring in patients with ferrous hip prostheses, this study evaluated whether megavoltage CT (MVCT) reduced interobserver variability.

METHODS

Two patients without hip prostheses and one patient (Patient 3) with hip prostheses were enrolled. Six radiation oncologists contoured prostate, rectum and bladder on kVCT (Patients 1 and 3), MRI (Patient 2) and MVCT images (Patient 3). MVCT was acquired with fine, normal and coarse modalities. Interobserver variability for each organ was analysed using conformity index (CI) and coefficient of variation (CV).

RESULTS

In patients without hip prostheses, CIs were higher in prostate contouring with MRI than with kVCT, indicating lower interobserver variability with MRI. Very slight variations were seen in rectum and bladder contouring. In the patient with hip prostheses (Patient 3), contouring on kVCT lowered CI and increased CV in the prostate, bladder and rectum. The differences were more marked in the prostate. Only fine modality MVCT reduced interobserver variability and only for the prostate.

CONCLUSION

Even though greater noise and less soft-tissue contrast increase contouring variability with MVCT than with kVCT, lack of artefacts on MVCT could provide better image definition by this modality in hip prosthesis patients in whom MRI is precluded.

ADVANCES IN KNOWLEDGE

We recommend the fine modality MVCT for contouring hip prostheses patients.

Prospective assessment of deep inspiration breath-hold using 3-dimensional surface tracking for irradiation of left-sided breast cancer

PURPOSE

Deep inspiration breath hold (DIBH) is used to decrease cardiac irradiation during radiation therapy (RT) for breast cancer. The patients most likely to benefit and the impact on treatment time remain largely unknown. We sought to identify predictors for the use of DIBH and to quantify differences in dosimetry and treatment time using a prospective registry.

METHODS AND MATERIALS

A total of 150 patients with left breast cancer were enrolled. All patients were simulated with both free breathing (FB) and DIBH. RT was delivered by either modality. Alternate scans were planned with use of deformable registration to include identical RT volumes. DIBH patients were monitored by a real-time surface tracking system, AlignRT (Vision RT, Ltd, London, United Kingdom). Baseline characteristics and treatment times were compared by Fisher exact test and Wilcoxon rank sum test. Dosimetric endpoints were analyzed by Wilcoxon signed rank test, and linear regression identified predictors for change in mean heart dose (∆MHD).

RESULTS

We treated 38 patients with FB and 110 with DIBH. FB patients were older, more likely to have heart and lung disease, and less likely to receive chemotherapy or immediate reconstruction (all P < .05). Treatment times were not significantly different, but DIBH patients had greater variability in times (P = .0002). Of 146 evaluable patients, DIBH resulted in >20 cGy improvement in MHD in 107 patients but a >20 cGy increase in MHD in 14. Both MHD and lung V20 were significantly lower in DIBH than in paired FB plans. On multivariate analysis, younger age (4.18 cGy per year; P < .0001), higher body mass index (6.06 cGy/kg/m(2); P = .0018), and greater change in lung volumes (130 cGy/L; P = .003) were associated with greater ∆MHD.

CONCLUSIONS

DIBH improves cardiac dosimetry without significantly impacting treatment time in most patients. Greater inspiratory lung volumes augment this benefit. Because the improvement with DIBH was not uniform, patients should be scanned with both FB and DIBH.

QuickStart radiotherapy: an inter-professional approach to expedite radiotherapy treatment in early breast cancer

Background and purpose

This study aims to develop an expedited radiotherapy (RT) process and evaluate its time savings in women requiring whole breast RT.

Material and methods

An inter-professional RT team streamlined the computed tomography (CT) simulation and treatment pathway for a ‘QuickStart’ process. Target delineation was performed by an advanced practice radiation therapist and approved by the radiation oncologist (RO) for planning. Automated breast planning software was used for treatment planning and standard quality checks were performed. To assess time savings, the initial 25 QuickStart patients were matched with women who underwent whole breast simulation on the same day (±3 days), treated using the conventional process.

Results

A total of 73 post-lumpectomy women were treated through the QuickStart process; the median consent-to-RT was 2 days (range: 0–13) and the mean CT simulation-to-RT treatment was 2 hours and 42 minutes (SD 0:30). In the subgroup analysis, QuickStart patients saved an average of 11 days from CT simulation-to-RT and had shorter median wait-times for both surgery/chemotherapy-to-RT (60 versus 38 days;p=0·002) and consultation-to-RT (7 versus 20 days;p&lt;0·001).

Conclusions

Through inter-professional team efforts and the application of automated planning software, we have achieved a process that significantly decreases patient wait-times while maintaining the quality of whole breast RT.