Dataset Information

Pretreatment Identification of Head and Neck Cancer Nodal Metastasis and Extranodal Extension Using Deep Learning Neural Networks.

ABSTRACT: Identification of nodal metastasis and tumor extranodal extension (ENE) is crucial for head and neck cancer management, but currently only can be diagnosed via postoperative pathology. Pretreatment, radiographic identification of ENE, in particular, has proven extremely difficult for clinicians, but would be greatly influential in guiding patient management. Here, we show that a deep learning convolutional neural network can be trained to identify nodal metastasis and ENE with excellent performance that surpasses what human clinicians have historically achieved. We trained a 3-dimensional convolutional neural network using a dataset of 2,875 CT-segmented lymph node samples with correlating pathology labels, cross-validated and fine-tuned on 124 samples, and conducted testing on a blinded test set of 131 samples. On the blinded test set, the model predicted ENE and nodal metastasis each with area under the receiver operating characteristic curve (AUC) of 0.91 (95%CI: 0.85-0.97). The model has the potential for use as a clinical decision-making tool to help guide head and neck cancer patient management.

SUBMITTER: Kann BH

PROVIDER: S-EPMC6145900 | biostudies-literature | 2018 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Pretreatment Identification of Head and Neck Cancer Nodal Metastasis and Extranodal Extension Using Deep Learning Neural Networks.

Kann Benjamin H BH Aneja Sanjay S Loganadane Gokoulakrichenane V GV Kelly Jacqueline R JR Smith Stephen M SM Decker Roy H RH Yu James B JB Park Henry S HS Yarbrough Wendell G WG Malhotra Ajay A Burtness Barbara A BA Husain Zain A ZA

Scientific reports 20180919 1

Identification of nodal metastasis and tumor extranodal extension (ENE) is crucial for head and neck cancer management, but currently only can be diagnosed via postoperative pathology. Pretreatment, radiographic identification of ENE, in particular, has proven extremely difficult for clinicians, but would be greatly influential in guiding patient management. Here, we show that a deep learning convolutional neural network can be trained to identify nodal metastasis and ENE with excellent performa ...[more]

PMID: 30232350

Similar Datasets

Project description:In radiation oncology, predicting patient risk stratification allows specialization of therapy intensification as well as selecting between systemic and regional treatments, all of which helps to improve patient outcome and quality of life. Deep learning offers an advantage over traditional radiomics for medical image processing by learning salient features from training data originating from multiple datasets. However, while their large capacity allows to combine high-level medical imaging data for outcome prediction, they lack generalization to be used across institutions. In this work, a pseudo-volumetric convolutional neural network with a deep preprocessor module and self-attention (PreSANet) is proposed for the prediction of distant metastasis, locoregional recurrence, and overall survival occurrence probabilities within the 10 year follow-up time frame for head and neck cancer patients with squamous cell carcinoma. The model is capable of processing multi-modal inputs of variable scan length, as well as integrating patient data in the prediction model. These proposed architectural features and additional modalities all serve to extract additional information from the available data when availability to additional samples is limited. This model was trained on the public Cancer Imaging Archive Head-Neck-PET-CT dataset consisting of 298 patients undergoing curative radio/chemo-radiotherapy and acquired from 4 different institutions. The model was further validated on an internal retrospective dataset with 371 patients acquired from one of the institutions in the training dataset. An extensive set of ablation experiments were performed to test the utility of the proposed model characteristics, achieving an AUROC of [Formula: see text], [Formula: see text] and [Formula: see text] for DM, LR and OS respectively on the public TCIA Head-Neck-PET-CT dataset. External validation was performed on a retrospective dataset with 371 patients, achieving [Formula: see text] AUROC in all outcomes. To test for model generalization across sites, a validation scheme consisting of single site-holdout and cross-validation combining both datasets was used. The mean accuracy across 4 institutions obtained was [Formula: see text], [Formula: see text] and [Formula: see text] for DM, LR and OS respectively. The proposed model demonstrates an effective method for tumor outcome prediction for multi-site, multi-modal combining both volumetric data and structured patient clinical data.

Project description:BackgroundLymph node count (LNC) from neck dissection has been associated with undernutrition and survival in head and neck squamous cell carcinoma (HNSCC). As local components of the immune system, cervical lymph nodes may reflect anti-tumor immune status. This study investigates the relationship between decreased LNC, formation of tertiary lymphoid structures (TLS), and primary tumor infiltration by lymphocytes in undernourished patients.MethodsA matched-cohort study was conducted in a tertiary medical center, where neck dissection quality was standardized for a total of 384 subjects that were evaluated. Six head and neck cancer patients that underwent primary surgery including neck dissection with low LNC and BMI (low BMI < 23, low LNC ≤ 5.6 per neck level) were matched by stage, p16 status, and subsite to 16 patients with normal BMI and high LNC. Multiplexed immunohistochemistry was used to evaluate the tumor-infiltrating lymphocytes and the number and quality of TLS within primary tumors. Whole primary cancers underwent automated analysis and counting of leukocytes after multiplexed immunohistochemistry staining of tumor slides. A head and neck pathologist blindly scored the number and maturity of TLS. Descriptive statistics were used to analyze outcomes.ResultsThe patients with low BMI and low LNC had significantly fewer CD3 (p = 0.0136), CD8 (p = 0.0003), and CD20 (p = 0.0334) cells in their primary tumors compared to patients with normal BMI and LNC. The low BMI low LNC patients also had fewer mature TLS (0.83/tumor) in their primary cancers compared to patients with normal BMI and high LNC (5.4/tumor) and also had greater than fourfold lower mature TLS density (TLS per μm2 mean) (6.34 × 10-9 vs. 2.82 × 10-8), with significantly worsened survival relative to patients with low BMI and normal LNC and patients with normal BMI.ConclusionLow LNC predicts worsened survival only in low BMI HNSCC patients with non-HPV related tumors and in these patients is associated with markers of immunosuppression such as fewer tumor-infiltrating CD8+ T-cells, CD20+ cells, and fewer TLS in primary cancers compared to matched normal BMI patients with high LNC.

Dataset Information

Pretreatment Identification of Head and Neck Cancer Nodal Metastasis and Extranodal Extension Using Deep Learning Neural Networks.

Publications

Pretreatment Identification of Head and Neck Cancer Nodal Metastasis and Extranodal Extension Using Deep Learning Neural Networks.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets