Contouring workload in adjuvant breast cancer radiotherapy

How Long Does Contouring Really Take? Results of the Royal College of Radiologists Contouring Surveys

AIMS

The success and safety of modern radiotherapy relies on accurate contouring. Understanding the time taken to complete radiotherapy contours is critical to informing workforce planning and, in the context of a workforce shortfall, advocating for investment in technology and multi-professional skills mix. We aimed to quantify the time taken to delineate target volumes for radical radiotherapy.

MATERIALS AND METHODS

The Royal College of Radiologists circulated two electronic surveys via email to all clinical oncology consultants in the UK. The individual case survey requested anonymous data regarding the next five patients contoured for radical radiotherapy. The second survey collected data on respondents' usual practice in radiotherapy contouring.

RESULTS

The median time to contour one radiotherapy case was 85 minutes (IQR = 50-131 minutes). Marked variability between and within tumour sites was evident: paediatric cancers took the most time (median = 210 minutes, IQR = 87.5 minutes), followed by head and neck and gynaecological cancers (median = 120 minutes, IQR = 71 and 72.5 minutes respectively). Breast cancer contouring required the least time (median = 43 minutes, IQR = 60 minutes). Radiotherapy technique, inclusion of nodes and 4D CT planning were associated with longer contouring times. A non-medical professional was involved in contouring in 65% of cases, but clinical oncology consultants were involved in target volume delineation in 90% of cases, and OARs in 74%. Peer review took place in 46% of cases with 56% of consultants reporting no time for peer review in their job plan.

CONCLUSION

Contouring for radical radiotherapy is complex and time-consuming, and despite increasing involvement of non-medical professionals, clinical oncology consultants remain the primary practitioners. Peer review practice is variable and time is often a limiting factor. Many factors influence the time required for contouring, and departments should take these factors and the need for peer-review into account when developing job plans.

Contour similarity and its implication on inverse prostate SBRT treatment planning

PURPOSE

Success of auto-segmentation is measured by the similarity between auto and manual contours that is often quantified by Dice coefficient (DC). The dosimetric impact of contour variability on inverse planning has been rarely reported. The main aim of this study is to investigate whether automatically generated organs-at-risk (OARs) could be used in inverse prostate stereotactic body radiation therapy (SBRT) planning and whether the dosimetric parameters are still clinically acceptable after radiation oncologists modify the OARs.

METHODS AND MATERIALS

Planning computed tomography images from 10 patients treated with SBRT for prostate cancer were selected and automatically segmented by commercially available atlas-based software. The automatically generated OAR contours were compared with the manually drawn contours. Two volumetric modulated arc therapy (VMAT) plans, autoRec-VMAT (where only automatically generated rectums were used in optimization) and autoAll-VMAT (where automatically generated OARs were used in inverse optimization) were generated. Dosimetric parameters based on the manually drawn PTV and OARs were compared with the clinically approved plans.

RESULTS

The DCs for the rectum contours varied from 0.55 to 0.74 with a mean value of 0.665. Differences of D₉₅ of the PTV between autoRec-VMAT and manu-VMAT plans varied from 0.03% to -2.85% with a mean value of -0.64%. Differences of D_0.03cc of manual rectum between the two plans varied from -0.86% to 9.94% with a mean value of 2.71%. D₉₅ of PTV between autoAll-VMAT and manu-VMAT plans varied from 0.28% to -2.9% with a mean value -0.83%. Differences of D_0.03cc of manual rectum between the two plans varied from -0.76% to 6.72% with a mean value of 2.62%.

CONCLUSION

Our study implies that it is possible to use unedited automatically generated OARs to perform initial inverse prostate SBRT planning. After radiation oncologists modify/approve the OARs, the plan qualities based on the manually drawn OARs are still clinically acceptable, and a re-optimization may not be needed.

A survey of medical dosimetrists' perceptions of efficiency and consistency of auto-contouring software

Although auto-contouring methods were created to reduce the workload for the radiation oncology team, concern lies in whether auto-contouring can improve efficiency regarding generated contours of a treatment plan. Researchers have measured differences between auto-contouring algorithms and manual contour methods specific to the contouring of organs at risk (OAR). The problem lies in the paucity of literature specific to perceptions of auto-contouring and the impact on workflow efficiency. The purpose of this study was to measure medical dosimetrists' perceptions of how auto-contouring software impacts the treatment planning process. To measure perceptions, researchers surveyed medical dosimetrists about their perspectives on consistency and efficiency of auto-contouring during treatment planning. A (Qualtrics, Provo, UT) survey was created based on the 2 research questions in this study. The survey was distributed through email to 2598 full members of the American Association of Medical Dosimetrists (AAMD) who were certified by the MDCB; mostly medical dosimetrists but also included a small group of medical physicists. The email open rate was 39% (1024/2598) but the response rate for those who read the email was only 8.4% (86/1024). Of the survey respondents, 67% (59/86) used auto-contouring software; thus, eligible to complete the remainder of the survey. Majority of participants agreed that auto-contouring software decreases time spent contouring per patient; however, most agreed that manual contouring is more efficient. Therefore, it was inferred that a combination of both auto and manual contouring have an impact on workload efficiency.

Automatic Segmentation of Clinical Target Volume and Organs-at-Risk for Breast Conservative Radiotherapy Using a Convolutional Neural Network

Objective

Delineation of clinical target volume (CTV) and organs at risk (OARs) is important for radiotherapy but is time-consuming. We trained and evaluated a U-ResNet model to provide fast and consistent auto-segmentation.

Methods

We collected 160 patients’ CT scans with breast cancer who underwent breast-conserving surgery (BCS) and were treated with radiotherapy. CTV and OARs were delineated manually and were used for model training. The dice similarity coefficient (DSC) and 95th percentile Hausdorff distance (95HD) were used to assess the performance of our model. CTV and OARs were randomly selected as ground truth (GT) masks, and artificial intelligence (AI) masks were generated by the proposed model. Two clinicians randomly compared CTV score differences of the contour. The consistency between two clinicians was tested. Time cost for auto-delineation was evaluated.

Results

The mean DSC values of the proposed method were 0.94, 0.95, 0.94, 0.96, 0.96 and 0.93 for breast CTV, contralateral breast, heart, right lung, left lung and spinal cord, respectively. The mean 95HD values were 4.31mm, 3.59mm, 4.86mm, 3.18mm, 2.79mm and 4.37mm for the above structures, respectively. The average CTV scores for AI and GT were 2.89 versus 2.92 when evaluated by oncologist A (P=0.612), and 2.75 versus 2.83 by oncologist B (P=0.213), with no statistically significant differences. The consistency between two clinicians was poor (kappa=0.282). The time for auto-segmentation of CTV and OARs was 10.03 s.

Conclusion

Our proposed model (U-ResNet) can improve the efficiency and accuracy of delineation compared with U-Net, performing equally well with the segmentation generated by oncologists.

A rapid review of influential factors and appraised solutions on organ delineation uncertainties reduction in radiotherapy

Background and purpose.Accurate volume delineation plays an essential role in radiotherapy. Contouring is a potential source of uncertainties in radiotherapy treatment planning that could affect treatment outcomes. Therefore, reducing the degree of contouring uncertainties is crucial. The role of utilized imaging modality in the organ delineation uncertainties has been investigated. This systematic review explores the influential factors on inter-and intra-observer uncertainties of target volume and organs at risk (OARs) delineation focusing on the used imaging modality for these uncertainties reduction and the reported subsequent histopathology and follow-up assessment.Methods and materials.An inclusive search strategy has been conducted to query the available online databases (Scopus, Google Scholar, PubMed, and Medline). 'Organ at risk', 'target', 'delineation', 'uncertainties', 'radiotherapy' and their relevant terms were utilized using every database searching syntax. Final article extraction was performed following Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline. Included studies were limited to the ones published in English between 1995 and 2020 and that just deal with computed tomography (CT) and magnetic resonance imaging (MRI) modalities.Results.A total of 923 studies were screened and 78 were included of which 31 related to the prostate 20 to the breast, 18 to the head and neck, and 9 to the brain tumor site. 98% of the extracted studies performed volumetric analysis. Only 24% of the publications reported the dose deviations resulted from variation in volume delineation Also, heterogeneity in studied populations and reported geometric and volumetric parameters were identified such that quantitative synthesis was not appropriate.Conclusion.This review highlightes the inter- and intra-observer variations that could lead to contouring uncertainties and impede tumor control in radiotherapy. For improving volume delineation and reducing inter-observer variability, the implementation of well structured training programs, homogeneity in following consensus and guidelines, reliable ground truth selection, and proper imaging modality utilization could be clinically beneficial.

Is prone free breathing better than supine deep inspiration breath-hold for left whole-breast radiotherapy? A dosimetric analysis

PURPOSE

The advantage of prone setup compared with supine for left-breast radiotherapy is controversial. We evaluate the dosimetric gain of prone setup and aim to identify predictors of the gain.

METHODS

Left-sided breast cancer patients who had dual computed tomography (CT) planning in prone free breathing (FB) and supine deep inspiration breath-hold (DiBH) were retrospectively identified. Radiation doses to heart, lungs, breasts, and tumor bed were evaluated using the recently developed mean absolute dose deviation (MADD). MADD measures how widely the dose delivered to a structure deviates from a reference dose specified for the structure. A penalty score was computed for every treatment plan as a weighted sum of the MADDs normalized to the breast prescribed dose. Changes in penalty scores when switching from supine to prone were assessed by paired t-tests and by the number of patients with a reduction of the penalty score (i.e., gain). Robust linear regression and fractional polynomials were used to correlate patients' characteristics and their respective penalty scores.

RESULTS

Among 116 patients identified with dual CT planning, the prone setup, compared with supine, was associated with a dosimetric gain in 72 (62.1%, 95% CI: 52.6-70.9%). The most significant predictors of a gain with the prone setup were the breast depth prone/supine ratio (>1.6), breast depth difference (>31 mm), prone breast depth (>77 mm), and breast volume (>282 mL).

CONCLUSION

Prone compared with supine DiBH was associated with a dosimetric gain in 62.1% of our left-sided breast cancer patients. High pendulousness and moderately large breast predicted for the gain.

Automatic contouring system for cervical cancer using convolutional neural networks

PURPOSE

To develop a tool for the automatic contouring of clinical treatment volumes (CTVs) and normal tissues for radiotherapy treatment planning in cervical cancer patients.

METHODS

An auto-contouring tool based on convolutional neural networks (CNN) was developed to delineate three cervical CTVs and 11 normal structures (seven OARs, four bony structures) in cervical cancer treatment for use with the Radiation Planning Assistant, a web-based automatic plan generation system. A total of 2254 retrospective clinical computed tomography (CT) scans from a single cancer center and 210 CT scans from a segmentation challenge were used to train and validate the CNN-based auto-contouring tool. The accuracy of the tool was evaluated by calculating the Sørensen-dice similarity coefficient (DSC) and mean surface and Hausdorff distances between the automatically generated contours and physician-drawn contours on 140 internal CT scans. A radiation oncologist scored the automatically generated contours on 30 external CT scans from three South African hospitals.

RESULTS

The average DSC, mean surface distance, and Hausdorff distance of our CNN-based tool were 0.86/0.19 cm/2.02 cm for the primary CTV, 0.81/0.21 cm/2.09 cm for the nodal CTV, 0.76/0.27 cm/2.00 cm for the PAN CTV, 0.89/0.11 cm/1.07 cm for the bladder, 0.81/0.18 cm/1.66 cm for the rectum, 0.90/0.06 cm/0.65 cm for the spinal cord, 0.94/0.06 cm/0.60 cm for the left femur, 0.93/0.07 cm/0.66 cm for the right femur, 0.94/0.08 cm/0.76 cm for the left kidney, 0.95/0.07 cm/0.84 cm for the right kidney, 0.93/0.05 cm/1.06 cm for the pelvic bone, 0.91/0.07 cm/1.25 cm for the sacrum, 0.91/0.07 cm/0.53 cm for the L4 vertebral body, and 0.90/0.08 cm/0.68 cm for the L5 vertebral bodies. On average, 80% of the CTVs, 97% of the organ at risk, and 98% of the bony structure contours in the external test dataset were clinically acceptable based on physician review.

CONCLUSIONS

Our CNN-based auto-contouring tool performed well on both internal and external datasets and had a high rate of clinical acceptability.

Analysis of Geometric Performance and Dosimetric Impact of Using Automatic Contour Segmentation for Radiotherapy Planning

Purpose: To analyze geometric discrepancy and dosimetric impact in using contours generated by auto-segmentation (AS) against manually segmented (MS) clinical contours. Methods: A 48-subject prostate atlas was created and another 15 patients were used for testing. Contours were generated using a commercial atlas-based segmentation tool and compared to their clinical MS counterparts. The geometric correlation was evaluated using the Dice similarity coefficient (DSC) and Hausdorff distance (HD). Dosimetric relevance was evaluated for a subset of patients by assessing the DVH differences derived by optimizing plan dose using the AS and MS contours, respectively, and evaluating with respect to each. A paired t-test was employed for statistical comparison. The discrepancy in plan quality with respect to clinical dosimetric endpoints was evaluated. The analysis was repeated for head/neck (HN) with a 31-subject atlas and 15 test cases. Results: Dice agreement between AS and MS differed significantly across structures: from (L:0.92/R: 0.91) for the femoral heads to seminal vesical of 0.38 in the prostate cohort, and from 0.98 for the brain, to 0.36 for the chiasm of the HN group. Despite the geometric disagreement, the paired t-tests showed the lack of statistical evidence for systematic differences in dosimetric plan quality yielded by the AS and MS approach for the prostate cohort. In HN cases, statistically significant differences in dosimetric endpoints were observed in structures with small volumes or elongated shapes such as cord (p = 0.01) and esophagus (p = 0.04). The largest absolute dose difference of 11 Gy was seen in the mean pharynx dose. Conclusion: Varying AS performance among structures suggests a differential approach of using AS on a subset of structures and focus MS on the rest. The discrepancy between geometric and dosimetric-end-point driven evaluation also indicates the clinical utility of AS contours in optimization and evaluating plan quality despite of suboptimal geometrical accuracy.

Generalization vs. Specificity: In Which Cases Should a Clinic Train its Own Segmentation Models?

As artificial intelligence for image segmentation becomes increasingly available, the question whether these solutions generalize between different hospitals and geographies arises. The present study addresses this question by comparing multi-institutional models to site-specific models. Using CT data sets from four clinics for organs-at-risk of the female breast, female pelvis and male pelvis, we differentiate between the effect from population differences and differences in clinical practice. Our study, thus, provides guidelines to hospitals, in which case the training of a custom, hospital-specific deep neural network is to be advised and when a network provided by a third-party can be used. The results show that for the organs of the female pelvis and the heart the segmentation quality is influenced solely on bases of the training set size, while the patient population variability affects the female breast segmentation quality above the effect of the training set size. In the comparison of site-specific contours on the male pelvis, we see that for a sufficiently large data set size, a custom, hospital-specific model outperforms a multi-institutional one on some of the organs. However, for small hospital-specific data sets a multi-institutional model provides the better segmentation quality.

Systematic review of educational interventions to improve contouring in radiotherapy

BACKGROUND AND PURPOSE

Contouring is a critical step in the radiotherapy process, but there is limited research on how to teach it and no consensus about the best method. We summarize the current evidence regarding improvement of contouring skills.

METHODS AND MATERIALS

Comprehensive literature search of the Pubmed-MEDLINE database, EMBASE database and Cochrane Library to identify relevant studies (independently examined by two investigators) that included baseline contouring followed by a re-contouring assessment after an educational intervention.

RESULTS

598 papers were identified. 16 studies met the inclusion criteria representing 370 participants (average number of participants per study of 23; range (4-141). Regarding the teaching methodology, 5/16 used onsite courses, 8/16 online courses, and 2/16 used blended learning. Study quality was heterogenous. There were only 3 randomized studies and only 3 analyzed the dosimetric impact of improving contouring homogeneity. Dice similarity coefficient was the most common evaluation metric (7/16), and in all these studies at least some contours improved significantly post-intervention. The time frame for evaluating the learning effect of the teaching intervention was almost exclusively short-time, with only one study evaluating the long-term utility of the educational program beyond 6 months.

CONCLUSION

The literature on educational interventions designed to improve contouring performance is limited and heterogenous. Onsite, online and blended learning courses have all been shown to be helpful, however, sample sizes are small and impact assessment is almost exclusively short-term and typically does not take into account the effect on treatment planning. The most effective teaching methodology/format is unknown and impact on daily clinical practice is uncertain.

Automatic detection of contouring errors using convolutional neural networks

PURPOSE

To develop a head and neck normal structures autocontouring tool that could be used to automatically detect the errors in autocontours from a clinically validated autocontouring tool.

METHODS

An autocontouring tool based on convolutional neural networks (CNN) was developed for 16 normal structures of the head and neck and tested to identify the contour errors from a clinically validated multiatlas-based autocontouring system (MACS). The computed tomography (CT) scans and clinical contours from 3495 patients were semiautomatically curated and used to train and validate the CNN-based autocontouring tool. The final accuracy of the tool was evaluated by calculating the Sørensen-Dice similarity coefficients (DSC) and Hausdorff distances between the automatically generated contours and physician-drawn contours on 174 internal and 24 external CT scans. Lastly, the CNN-based tool was evaluated on 60 patients' CT scans to investigate the possibility to detect contouring failures. The contouring failures on these patients were classified as either minor or major errors. The criteria to detect contouring errors were determined by analyzing the DSC between the CNN- and MACS-based contours under two independent scenarios: (a) contours with minor errors are clinically acceptable and (b) contours with minor errors are clinically unacceptable.

RESULTS

The average DSC and Hausdorff distance of our CNN-based tool was 98.4%/1.23 cm for brain, 89.1%/0.42 cm for eyes, 86.8%/1.28 cm for mandible, 86.4%/0.88 cm for brainstem, 83.4%/0.71 cm for spinal cord, 82.7%/1.37 cm for parotids, 80.7%/1.08 cm for esophagus, 71.7%/0.39 cm for lenses, 68.6%/0.72 for optic nerves, 66.4%/0.46 cm for cochleas, and 40.7%/0.96 cm for optic chiasm. With the error detection tool, the proportions of the clinically unacceptable MACS contours that were correctly detected were 0.99/0.80 on average except for the optic chiasm, when contours with minor errors are clinically acceptable/unacceptable, respectively. The proportions of the clinically acceptable MACS contours that were correctly detected were 0.81/0.60 on average except for the optic chiasm, when contours with minor errors are clinically acceptable/unacceptable, respectively.

CONCLUSION

Our CNN-based autocontouring tool performed well on both the publically available and the internal datasets. Furthermore, our results show that CNN-based algorithms are able to identify ill-defined contours from a clinically validated and used multiatlas-based autocontouring tool. Therefore, our CNN-based tool can effectively perform automatic verification of MACS contours.

A Full-Image Deep Segmenter for CT Images in Breast Cancer Radiotherapy Treatment

Radiation therapy is one of the key cancer treatment options. To avoid adverse effects in the healthy tissue, the treatment plan needs to be based on accurate anatomical models of the patient. In this work, an automatic segmentation solution for both female breasts and the heart is constructed using deep learning. Our newly developed deep neural networks perform better than the current state-of-the-art neural networks while improving inference speed by an order of magnitude. While manual segmentation by clinicians takes around 20 min, our automatic segmentation takes less than a second with an average of 3 min manual correction time. Thus, our proposed solution can have a huge impact on the workload of clinical staff and on the standardization of care.