Multicenter evaluation of different target volume delineation concepts in pediatric Hodgkin's lymphoma. A case study

Deep Learning Algorithm for Auto-Delineation of High-Risk Oropharyngeal Clinical Target Volumes With Built-In Dice Similarity Coefficient Parameter Optimization Function

PURPOSE

Automating and standardizing the contouring of clinical target volumes (CTVs) can reduce interphysician variability, which is one of the largest sources of uncertainty in head and neck radiation therapy. In addition to using uniform margin expansions to auto-delineate high-risk CTVs, very little work has been performed to provide patient- and disease-specific high-risk CTVs. The aim of the present study was to develop a deep neural network for the auto-delineation of high-risk CTVs.

METHODS AND MATERIALS

Fifty-two oropharyngeal cancer patients were selected for the present study. All patients were treated at The University of Texas MD Anderson Cancer Center from January 2006 to August 2010 and had previously contoured gross tumor volumes and CTVs. We developed a deep learning algorithm using deep auto-encoders to identify physician contouring patterns at our institution. These models use distance map information from surrounding anatomic structures and the gross tumor volume as input parameters and conduct voxel-based classification to identify voxels that are part of the high-risk CTV. In addition, we developed a novel probability threshold selection function, based on the Dice similarity coefficient (DSC), to improve the generalization of the predicted volumes. The DSC-based function is implemented during an inner cross-validation loop, and probability thresholds are selected a priori during model parameter optimization. We performed a volumetric comparison between the predicted and manually contoured volumes to assess our model.

RESULTS

The predicted volumes had a median DSC value of 0.81 (range 0.62-0.90), median mean surface distance of 2.8 mm (range 1.6-5.5), and median 95th Hausdorff distance of 7.5 mm (range 4.7-17.9) when comparing our predicted high-risk CTVs with the physician manual contours.

CONCLUSIONS

These predicted high-risk CTVs provided close agreement to the ground-truth compared with current interobserver variability. The predicted contours could be implemented clinically, with only minor or no changes.

A dose based approach for evaluation of inter-observer variations in target delineation

BACKGROUND AND PURPOSE

Substantial inter-observer variations in target delineation have been presented previously. Target delineation for paediatric cases is difficult due to the small number of children, the variation in paediatric targets, the number of study protocols, and the individual patient's specific needs and demands. Uncertainties in target delineation might lead to under-dosage or over-dosage. The aim of this work is to apply the concept of a consensus volume and good quality treatment plans to visualise and quantify inter-observer target delineation variations in dosimetric terms in addition to conventional geometrically based volume concordance indices.

MATERIAL AND METHODS

Two paediatric cases were used to demonstrate the potential of adding dose metrics when evaluating target delineation diversity; Hodgkin's disease (case 1) and rhabdomyosarcoma of the parotid gland (case 2). The variability in target delineation (PTV delineations) between six centres was quantified using the generalised conformity index, CIgen, generated for volume overlap. The STAPLE algorithm, as implemented in CERR, was used for both cases to derive a consensus volumes. STAPLE is a probabilistic estimate of the true volume generated from all observers. Dose distributions created by each centre for the original target volumes were then applied to this consensus volume.

RESULTS

A considerable variation in target segmentation was seen in both cases. For case 1 the variation was 374-960 cm3 (average 669 cm3) and for case 2; 65-126 cm3 (average 109 cm3). CIgen were 0.53 and 0.70, respectively. The DVHs in absolute volume displayed for the delineated target volume as well as for the consensus volume adds information on both "compliant" target volumes as well as outliers which are hidden with just the use of concordance indices.

CONCLUSIONS

The DVHs in absolute volume add valuable and easily understood information to various indices for evaluating uniformity in target delineation.

Modelling tumour volume variations in head and neck cancer: contribution of magnetic resonance imaging for patients undergoing induction chemotherapy

Primary tumour volume evaluation has predictive value for estimating survival outcomes. Using volumetric data acquired by MRI in patients undergoing induction chemotherapy (IC) these outcomes were estimated before the radiotherapy course in head and neck cancer (HNC) patients. MRI performed before and after IC in 36 locally advanced HNC patients were analysed to measure primary tumour volume. The two volumes were correlated using the linear-log ratio (LLR) between the volume in the first MRI and the volume in the second. Cox's proportional hazards models (CPHM) were defined for loco-regional control (LRC), disease-free survival (DFS) and overall survival (OS). Strict evaluation of the influence of volume delineation uncertainties on prediction of final outcomes has been defined. LLR showed good predictive value for all survival outcomes in CPHM. Predictive models for LRC and DFS at 24 months showed optimal discrimination and prediction capability. Evaluation of primary tumour volume variations in HNC after IC provides an example of modelling that can be easily used even for other adaptive treatment approaches. A complete assessment of uncertainties in covariates required for running models is a prerequisite to create reliable clinically models.

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.

Interobserver variability in clinical target volume delineation for primary mediastinal B-cell lymphoma

PURPOSE

The purpose of this study was to evaluate interobserver variability among radiation oncologists with experience in the field of lymphoma radiation therapy in the delineation of clinical target volume (CTV) in a challenging case of primary mediastinal B-cell lymphoma.

METHODS AND MATERIALS

Ten experienced radiation oncologists were invited to a 1-day contouring session. The case of a 56-year-old man with primary mediastinal B-cell lymphoma with complete metabolic response after chemotherapy was chosen as the sample for the study. A brief presentation of his clinical history was given, together with guidelines for contouring. The 10 CTVs obtained were then compared in terms of variation in total volume and in craniocaudal, laterolateral, and anteroposterior diameters. The CTV with the best Dice similarity coefficient (DSC) between the union of all 10 CTVs and the individual CTV was considered the reference CTV, and the DSC and the Hausdorff distance (HD) for each volume compared with the reference CTV were then calculated.

RESULTS

A significant variability was found in total volume (mean, 498.3 cm(3); range, 181.8-1003 cm(3)) and craniocaudal (median, 144.7 mm; range, 80.6-159 mm), laterolateral (median, 133.5 mm; range, 83.7-149.5 mm), and anteroposterior diameters (median, 136.2 mm; range, 84-150.5 mm). Analysis of the DSC and the HD showed a mean DSC of 0.53 (range, 0.31-0.74) and a mean HD of 6.4 cm (range, 1.8-14.8 cm).

CONCLUSIONS

Results of this study strongly indicate the need to develop and share appropriate contouring guidelines among experts and suggest the promotion of specific educational activities to improve radiation therapy quality in both clinical trials and routine clinical practice.

Delineation of the larynx as organ at risk in radiotherapy: a contouring course within "Rete Oncologica Piemonte-Valle d'Aosta" network to reduce inter- and intraobserver variability

AIMS

To evaluate the usefulness of a contouring course in reducing inter- and intraobserver variability in the definition of the larynx as organ at risk (OAR).

METHODS

Within the "Rete Oncologica Piemonte-Valle d'Aosta" network, a contouring course focusing on larynx delineation was proposed. Twenty-six radiotherapist technicians (RTTs) experienced in delineating OARs were asked to contour larynx before and after the training. An expert radiation oncologist defined the reference volume for educational purpose. The contoured volumes obtained before and after the course were compared using descriptive statistics (mean value, standard deviation-SD, and coefficient of variation-COV) of volumes and maximum diameters. Conformity index (CI), dice coefficient (DC), and percentage of overlap were used to evaluate the spatial accuracy of the different volumes compared to the reference. Further analysis regarding the variation in the centre of mass (COM) displacement was performed.

RESULTS

The mean volume was 40.4 cm³ before and 65.9 cm³ after the course, approaching the reference value. Mean anteroposterior, laterolateral, and craniocaudal diameters improved, getting each closer to the reference. Moreover, the COM moved approaching reference coordinates. Mean percentage of intersection and DC strongly increased after the course, rising from 57.76 to 93.83 % and from 0.68 to 0.89, respectively. CI enhanced from 0.06 to 0.31.

CONCLUSIONS

This study shows an improvement in larynx definition after the contouring course with lower interobserver variability and major consistency compared to the reference volume. Other specific educational activities may further increase the quality of radiation therapy contouring in this setting.

Does ultrasound provide any added value in breast contouring for radiotherapy after conserving surgery for cancer?

BACKGROUND

Whole breast irradiation after conserving surgery for breast cancer requires precise definition of the target volume. The standard approach uses computed tomography (CT) images. However, since fatty breast and non-breast tissues have similar electronic densities, difficulties in differentiating between them hamper breast volume delineation. To overcome this limitation the breast contour is defined by palpation and then radio-opaque wire is put around it before the CT scan. To optimize assessment of breast margins in the cranial, caudal, medial, lateral and posterior directions, the present study evaluated palpation and CT and determined whether ultrasound (US) provided any added value.

METHODS

Twenty consecutive patients were enrolled after they had provided informed consent to participating in this prospective study which was approved by the Regional Public Health Ethics Committee. Palpation and US defined breast margins and each contour was marked and outlined with a fine plastic wire. Breasts were then contoured on axial CT images using the breast window width (WW) and window level (WL) (401 and 750 Hounsfield Units -HU- respectively), at which setting the plastic wires were invisible. Then, the lung window function (WW 1601 HU; WL -300 HU) was inserted to visualize the plastic wires which were used as guidelines to contour the palpable and US breast volumes. As each wire had a different diameter, both volumes were easily defined on CT slices. Results were analyzed using descriptive statistics, percentage overlap and reproducibility measures (agreement and reliability).

RESULTS

Volumes: US gave the largest and palpation the smallest. Agreement was best between palpation and CT. Reliability was almost perfect in all correlations. Extensions: Cranial and posterior were highest with US and smallest with palpation. Agreement was best between palpation and CT in all extensions except the cranial. Since strong to almost perfect agreement emerged for all comparisons, reliability was high.

CONCLUSIONS

US may be useful in defining the cranial and posterior extensions, mainly when tumours are localized there. This study demonstrates that the now standard radio-opaque wires around the palpable breast may not be needed in breast contouring.

Radiotherapy of Hodgkin and Non-Hodgkin Lymphoma

Purpose:

To improve the contouring of clinical target volume for the radiotherapy of neck Hodgkin/non-Hodgkin lymphoma by localizing the prechemotherapy gross target volume onto the simulation computed tomography using [18F]-fluorodeoxyglucose positron emission tomography/computed tomography.

Material and Methods:

The gross target volume delineated on prechemotherapy [18F]-fluorodeoxyglucose positron emission tomography/computed tomography images was warped onto simulation computed tomography using deformable image registration. Fifteen patients with neck Hodgkin/non-Hodgkin lymphoma were analyzed. Quality of image registration was measured by computing the Dice similarity coefficient on warped organs at risk. Five radiation oncologists visually scored the localization of automatic gross target volume, ranking it from 1 (wrong) to 5 (excellent). Deformable registration was compared to rigid registration by computing the overlap index between the automatic gross target volume and the planned clinical target volume and quantifying the V95 coverage.

Results:

The Dice similarity coefficient was 0.80 ± 0.07 (median ± quartiles). The physicians’ survey had a median score equal to 4 (good). By comparing the rigid versus deformable registration, the overlap index increased from a factor of about 4 and the V95 (percentage of volume receiving the 95% of the prescribed dose) went from 0.84 ± 0.38 to 0.99 ± 0.10 (median ± quartiles).

Conclusion:

This study demonstrates the impact of using deformable registration between prechemotherapy [18F]-fluorodeoxyglucose positron emission tomography/computed tomography and simulation computed tomography, in order to automatically localize the gross target volume for radiotherapy treatment of patients with Hodgkin/non-Hodgkin lymphoma.

Novel radiotherapy techniques for involved-field and involved-node treatment of mediastinal Hodgkin lymphoma: when should they be considered and which questions remain open?

PURPOSE

Hodgkin lymphoma (HL) is a highly curable disease. Reducing late complications and second malignancies has become increasingly important. Radiotherapy target paradigms are currently changing and radiotherapy techniques are evolving rapidly.

DESIGN

This overview reports to what extent target volume reduction in involved-node (IN) and advanced radiotherapy techniques, such as intensity-modulated radiotherapy (IMRT) and proton therapy-compared with involved-field (IF) and 3D radiotherapy (3D-RT)- can reduce high doses to organs at risk (OAR) and examines the issues that still remain open.

RESULTS

Although no comparison of all available techniques on identical patient datasets exists, clear patterns emerge. Advanced dose-calculation algorithms (e.g., convolution-superposition/Monte Carlo) should be used in mediastinal HL. INRT consistently reduces treated volumes when compared with IFRT with the exact amount depending on the INRT definition. The number of patients that might significantly benefit from highly conformal techniques such as IMRT over 3D-RT regarding high-dose exposure to organs at risk (OAR) is smaller with INRT. The impact of larger volumes treated with low doses in advanced techniques is unclear. The type of IMRT used (static/rotational) is of minor importance. All advanced photon techniques result in similar potential benefits and disadvantages, therefore only the degree-of-modulation should be chosen based on individual treatment goals. Treatment in deep inspiration breath hold is being evaluated. Protons theoretically provide both excellent high-dose conformality and reduced integral dose.

CONCLUSION

Further reduction of treated volumes most effectively reduces OAR dose, most likely without disadvantages if the excellent control rates achieved currently are maintained. For both IFRT and INRT, the benefits of advanced radiotherapy techniques depend on the individual patient/target geometry. Their use should therefore be decided case by case with comparative treatment planning.

Interobserver variability of clinical target volume delineation in soft-tissue sarcomas

PURPOSE

The present observational study reports the results of a multi-institutional dummy-run designed to estimate the consistency of interobserver variability in clinical target volume delineation in two different cases of soft-tissue sarcomas in which postoperative and preoperative radiotherapy were prescribed, respectively. The purpose of this work was to quantify interobserver variability in routine clinical practice.

PATIENTS AND METHODS

Two different cases of soft-tissues sarcomas were chosen: a case of postoperative and a case of preoperative radiation therapy. Participating centres were requested to delineate clinical target volumes according to their experience in both cases. Descriptive statistic was calculated for each variable (volume, diameters) separately for two cases. Box-whiskers plots were used for presentation of clinical target volume. A Shapiro-Wilk's test was performed to evaluate the departures from normality distribution for each variable. The comparison between relative variations of diameters was evaluated using the Student's t test.

RESULTS

Several variations affecting both volumes and diameters were observed. Main variations were observed in the craniocaudal and laterolateral diameters. Each case showed similar dispersion, indicating a lack of reproducibility in volumes definition.

CONCLUSIONS

This observational study highlighted that, in the absence of specific instructions or guidelines, the interobserver variability can be significant both in postoperative and preoperative radiotherapy of soft-tissue sarcomas.