Overview of the HECKTOR Challenge at MICCAI 2020: Automatic Head and Neck Tumor Segmentation in PET/CT

Andrearczyk, Vincent; Oreiller, Valentin; Jreige, Mario; Vallières, Martin; Castelli, J.; Elhalawani, Hesham; Boughdad, Sarah; Prior, John O.; Depeursinge, Adrien

doi:10.1007/978-3-030-67194-5_1

Cited by 86 publications

(79 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, there has been increasing evidence suggesting the utility of applying deep learning for fully-automated OPC tumor auto-segmentation in various imaging modalities [14,[20][21][22]. PET-CT has recently shown excellent performance when used as inputs to deep learning models, partly due to the large and highly curated datasets provided by the HECKTOR Challenge [8]. While direct comparison of performance metrics between segmentation studies is often ill-advised, the HECKTOR Challenge offers a systematic method for directly compare segmentation methods with each other.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Head and Neck Cancer Primary Tumor Auto Segmentation using Model Ensembling of Deep Learning in PET-CT Images

Naser

Wahid

Dijk

et al. 2021

Preprint

View full text Add to dashboard Cite

Auto-segmentation of primary tumors in oropharyngeal cancer using PET/CT images is an unmet need that has the potential to improve radiation oncology workflows. In this study, we develop a series of deep learning models based on a 3D Residual Unet (ResUnet) architecture that can segment oropharyngeal tumors with high performance as demonstrated through internal and external validation of large-scale datasets (training size = 224 patients, testing size = 101 patients) as part of the 2021 HECKTOR Challenge. Specifically, we leverage ResUNet models with either 256 or 512 bottleneck layer channels that are able to demonstrate internal validation (10-fold cross-validation) mean Dice similarity coefficient (DSC) up to 0.771 and median 95% Hausdorff distance (95% HD) as low as 2.919 mm. We employ label fusion ensemble approaches, including Simultaneous Truth and Performance Level Estimation (STAPLE) and a voxel-level threshold approach based on majority voting (AVERAGE), to generate consensus segmentations on the test data by combining the segmentations produced through different trained cross-validation models. We demonstrate that our best performing ensembling approach (256 channels AVERAGE) achieves a mean DSC of 0.770 and median 95% HD of 3.143 mm through independent external validation on the test set. Concordance of internal and external validation results suggests our models are robust and can generalize well to unseen PET/CT data. We advocate that ResUNet models coupled to label fusion ensembling approaches are promising candidates for PET/CT oropharyngeal primary tumors auto-segmentation, with future investigations targeting the ideal combination of channel combinations and label fusion strategies to maximize seg-mentation performance.

show abstract

Section: Discussionmentioning

confidence: 99%

“…The data set used in this study, which was released through AIcrowd [7] for the HECKTOR Challenge at MICCAI 2021 [8], consists of co-registered 18…”

Section: Imaging Datamentioning

confidence: 99%

Head and Neck Cancer Primary Tumor Auto Segmentation using Model Ensembling of Deep Learning in PET-CT Images

Naser

Wahid

Dijk

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Data from 224 HNSCC patients from multiple institutions was provided in the 2021 HECKTOR Challenge [6,7] training set. Data for these patients included co-registered 18 F-FDG PET and CT scans, clinical data (Center ID, Gender, Age, TNM edition, chemotherapy status, TNM group, T-stage, N-stage, and M-stage), and ground truth manual segmentations of primary tumors derived from clinical experts.…”

Section: Methodsmentioning

confidence: 99%

Combining Tumor Segmentation Masks with PET/CT Images and Clinical Data in a Deep Learning Framework for Improved Prognostic Prediction in Head and Neck Squamous Cell Carcinoma

Wahid

Dede

et al. 2021

Preprint

View full text Add to dashboard Cite

PET/CT images provide a rich data source for clinical prediction models in head and neck squamous cell carcinoma (HNSCC). Deep learning models often use images in an end-to-end fashion with clinical data or no additional input for predictions. However, in the context of HNSCC, the tumor region of interest may be an informative prior in the generation of improved prediction performance. In this study, we utilize a deep learning framework based on a DenseNet architecture to combine PET images, CT images, primary tumor segmentation masks, and clinical data as separate channels to predict progression-free survival (PFS) in days for HNSCC patients. Through internal validation (10-fold cross-validation) based on a large set of training data provided by the 2021 HECKTOR Challenge, we achieve a mean C-index of 0.855 +- 0.060 and 0.650 +- 0.074 when observed events are and are not included in the C-index calculation, respectively. Ensemble approaches applied to cross-validation folds yield C-index values up to 0.698 in the independent test set (external validation). Importantly, the value of the added segmentation mask is underscored in both internal and external validation by an improvement of the C-index when compared to models that do not utilize the segmentation mask. These promising results highlight the utility of including segmentation masks as additional input channels in deep learning pipelines for clinical outcome prediction in HNSCC.

show abstract

“…Leading professional societies [e.g., MICCAI Society, Radiological Society of North America, American Association of Physicists in Medicine, Society of Nuclear Medicine and Molecular Imaging (SNMMI)] often organize international challenges for validation and comparison of competitive medical image analysis algorithms, leading to the creation of rankings and guidelines regarding the performance of these approaches under controlled conditions by using public image repositories (127). Several of these challenges were set up to tackle major issues facing quantitative PET imaging, including MICCAI Society challenges on PET image segmentation (128,129) and PET radiomics analysis (130), and the most recent SNMMI challenge on 177 Lu dosimetry based on quantitative SPECT imaging (131).…”

Section: Issues With Medical Imaging Challenges and Rankings Of Competitionsmentioning

confidence: 99%

Quantitative Molecular Positron Emission Tomography Imaging Using Advanced Deep Learning Techniques

Zaidi

Naqa

2021

Annu. Rev. Biomed. Eng.

View full text Add to dashboard Cite

The widespread availability of high-performance computing and the popularity of artificial intelligence (AI) with machine learning and deep learning (ML/DL) algorithms at the helm have stimulated the development of many applications involving the use of AI-based techniques in molecular imaging research. Applications reported in the literature encompass various areas, including innovative design concepts in positron emission tomography (PET) instrumentation, quantitative image reconstruction and analysis techniques, computer-aided detection and diagnosis, as well as modeling and prediction of outcomes. This review reflects the tremendous interest in quantitative molecular imaging using ML/DL techniques during the past decade, ranging from the basic principles of ML/DL techniques to the various steps required for obtaining quantitatively accurate PET data, including algorithms used to denoise or correct for physical degrading factors as well as to quantify tracer uptake and metabolic tumor volume for treatment monitoring or radiation therapy treatment planning and response prediction. This review also addresses future opportunities and current challenges facing the adoption of ML/DL approaches and their role in multimodality imaging. Expected final online publication date for the Annual Review of Biomedical Engineering, Volume 23 is June 2021. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

show abstract

Overview of the HECKTOR Challenge at MICCAI 2020: Automatic Head and Neck Tumor Segmentation in PET/CT

Cited by 86 publications

References 69 publications

Head and Neck Cancer Primary Tumor Auto Segmentation using Model Ensembling of Deep Learning in PET-CT Images

Head and Neck Cancer Primary Tumor Auto Segmentation using Model Ensembling of Deep Learning in PET-CT Images

Combining Tumor Segmentation Masks with PET/CT Images and Clinical Data in a Deep Learning Framework for Improved Prognostic Prediction in Head and Neck Squamous Cell Carcinoma

Quantitative Molecular Positron Emission Tomography Imaging Using Advanced Deep Learning Techniques

Contact Info

Product

Resources

About