Radiomics Analysis of Lymph Nodes with Esophageal Squamous Cell Carcinoma Based on Deep Learning.

Purpose: To assess the role of multiple radiomic features of lymph nodes in the preoperative prediction of lymph node metastasis (LNM) in patients with esophageal squamous cell carcinoma (ESCC). Methods: Three hundred eight patients with pathologically confirmed ESCC were retrospectively enrolled (training cohort, n = 216; test cohort, n = 92). We extracted 207 handcrafted radiomic features and 1000 deep radiomic features of lymph nodes from their computed tomography (CT) images. The t-test and least absolute shrinkage and selection operator (LASSO) were used to reduce the dimensions and select key features. Handcrafted radiomics, deep radiomics, and clinical features were combined to construct models. Models I (handcrafted radiomic features), II (Model I plus deep radiomic features), and III (Model II plus clinical features) were built using three machine learning methods: support vector machine (SVM), adaptive boosting (AdaBoost), and random forest (RF). The best model was compared with the results of two radiologists, and its performance was evaluated in terms of sensitivity, specificity, accuracy, area under the curve (AUC), and receiver operating characteristic (ROC) curve analysis. Results: No significant differences were observed between cohorts. Ten handcrafted and 12 deep radiomic features were selected from the extracted features (p < 0.05). Model III could discriminate between patients with and without LNM better than the diagnostic results of the two radiologists. Conclusion: The combination of handcrafted radiomic features, deep radiomic features, and clinical features could be used clinically to assess lymph node status in patients with ESCC.

Optimal Size Criterion for Malignant Lymph Nodes and a Novel Lymph Node Clinical Staging System for Unresectable Esophageal Squamous Cell Carcinoma: Evaluation by Multislice Spiral Computed Tomography.

Objectives: The current Chinese draft nodal clinical staging system for unresectable esophageal cancer is controversial. Our study aimed to propose a new diagnostic criterion for lymph node metastasis (LNM) detected by multislice spiral computed tomography (MSCT) in nonsurgically treated esophageal squamous cell carcinoma (ESCC) patients and then develop a novel lymph node (LN) clinical staging system for better individual prognostic prediction. Methods: The short-axis diameters of regional LNs were measured in 393 nonsurgical patients. Regional nodes were considered positive for malignancy if the nodal size exceeded the optimal size, which was determined by Kaplan-Meier survival analysis. The novel LN clinical staging system was then constructed using the LASSO model based on the relative prognostic importance of different LN stations. Validation cohort was included to confirm the prognostic performance. Results: Regional nodes were considered positive for malignancy if they were larger than 10 mm in the low cervical and upper thoracic segments, 7 mm in the middle thoracic segment, and 8 mm in the lower thoracic and celiac segments. Using the LASSO model, stations 2R, 3A, 7 and 16 were qualified in the model. Further analysis showed that our LN clinical staging system had better homogeneity, discriminatory ability and clinical value than the draft nodal staging system. Conclusions: Our results show that the new diagnostic criterion may improve the diagnostic value of MSCT in metastatic LNs. The novel LN clinical staging system can stratify nonsurgically treated ESCC patients into different risk groups, providing valuable information for decision making and outcome prediction.