Auto-delineation of oropharyngeal clinical target volumes using 3D convolutional neural networks

Abstract

Accurate clinical target volume (CTV) delineation is essential to ensure proper tumor coverage in radiation therapy. This is a particularly difficult task for head-and-neck cancer patients where detailed knowledge of the pathways of microscopic tumor spread is necessary. This paper proposes a solution to auto-segment these volumes in oropharyngeal cancer patients using a two-channel 3D U-Net architecture. The first channel feeds the network with the patient's CT image providing anatomical context, whereas the second channel provides the network with tumor location and morphological information.Radiation therapy simulation computer tomography scans and their corresponding manually delineated CTV and gross tumor volume (GTV) delineations from 285 oropharyngeal patients previously treated at MD Anderson Cancer Center were used in this study. CTV and GTV delineations underwent rigorous group peer-review prior to the start of treatment delivery. The convolutional network's parameters were fine-tuned using a training set of 210 patients using 3-fold cross-validation. During hyper-parameter selection, we use a score based on the overlap (dice similarity coefficient (DSC)) and missed volumes (false negative dice (FND)) to minimize any possible under-treatment. Three auto-delineated models were created to estimate tight, moderate, and wide CTV margin delineations.Predictions on our test set (75 patients) resulted in auto-delineations with high overlap and close surface distance agreement (DSC > 0.75 on 96% of cases for tight and moderate auto-delineation models and 97% of cases having mean surface distance ⩽ 5.0 mm) to the ground-truth. We found that applying a 5 mm uniform margin expansion to the auto-delineated CTVs would cover at least 90% of the physician CTV volumes for a large majority of patients; however, determination of appropriate margin expansions for auto-delineated CTVs merits further investigation.

Head and Neck Cancer Adaptive Radiation Therapy (ART): Conceptual Considerations for the Informed Clinician

“…For example, using the combination of a deep learning neural network and a shape representation model, Tong et al recently published a competitive algorithm that can delineate 9 OARs on a new scan under 10 seconds (20). Similarly, Cardenas et al (21,22)have shown that rapid CTV generation can be performed with machine learning approaches, obviating an often time-consuming step in target delineation. Although these results are promising, general consensus remains that automatic segmentation has yet to completely replace physician contouring, as manual checks and sometimes adjustments remain necessary(23); Voet et al (24), for example, showed that a commercial-softwareautosegmented CTV protocol delivered clinically meaningful undercoverage, which was not reflected decisively in similarity metric assessment, despite potential clinical risk if implemented without oversight.…”

Automatic detection of contouring errors using convolutional neural networks

“…The CNN algorithm was chosen to develop an autocontouring tool because other studies have shown that CNN‐based models outperform most other machine learning‐based and model‐based algorithms in contouring head and neck structures. Zhu et al, showed that their CNN‐based autocontouring algorithm for head and neck normal structures could achieve the equivalent performance with the best MICCAI 2015 challenge results with the atlas‐ and model‐based algorithms .…”

Deep Esophageal Clinical Target Volume Delineation Using Encoded 3D Spatial Context of Tumors, Lymph Nodes, and Organs At Risk

“…Comparison setup and metrics: We use 3-fold cross-validation, separated at the patient level, to evaluate performance of our approach and the competitor methods. We compare against setups using only the CT appearance information [14,15] and setups using the CT with binary GTV/LN masks [3]. Finally, we also compare against setups using the CT + GTV/LN SDMs, which does not consider the OARs.…”

“…1, CTV delineation depends on the radiation oncologist's visual judgment of both the appearance and the spatial configuration of the GTV, LNs, and OARs, suggesting that only considering the RTCT makes the problem ill-posed. Supporting this, Cardenas et al recently showed that considering the GTV and LN binary masks together with the RTCT can boost oropharyngeal CTV delineation performance [3]. However, the OARs were not considered in their work.…”

“…Since we are the first to tackle automated esophageal cancer CTV delineation, we compare against previous CTV delineation methods for other cancers [3,15], using the 3D PHNN as the delineation model. When comparing against pure appearance-based [15] and binary-mask-based [3] solutions, we show that our approach provides improvements of 10% and 3.8% in Dice score, respectively, with analogous improvements in Hausdorff distance (HD) and average surface distance (ASD). Moreover, we also show that PHNN is responsible for providing improvements of 1% in Dice score and 0.4mm reduction in ASD over a 3D U-Net model [4].…”