Dataset Information

A preoperative radiomics model for the identification of lymph node metastasis in patients with early-stage cervical squamous cell carcinoma.

ABSTRACT:

Objectives

To develop and validate a radiomics model for preoperative identification of lymph node metastasis (LNM) in patients with early-stage cervical squamous cell carcinoma (CSCC).

Methods

Total of 190 eligible patients were randomly divided into training (n = 100) and validation (n = 90) cohorts. Handcrafted features and deep-learning features were extracted from T2W fat suppression images. The minimum redundancy maximum relevance algorithm and LASSO regression with 10-fold cross-validation were used for key features selection. A radiomics model that incorporated the handcrafted-signature, deep-signature, and squamous cell carcinoma antigen (SCC-Ag) levels was developed by logistic regression. The model performance was assessed and validated with respect to its calibration, discrimination, and clinical usefulness.

Results

Three handcrafted features and three deep-learning features were selected and used to build handcrafted- and deep-signature. The model, which incorporated the handcrafted-signature, deep-signature, and SCC-Ag, showed satisfactory calibration and discrimination in the training cohort (AUC: 0.852, 95% CI: 0.761-0.943) and the validation cohort (AUC: 0.815, 95% CI: 0.711-0.919). Decision curve analysis indicated the clinical usefulness of the radiomics model. The radiomics model yielded greater AUCs than either the radiomics signature (AUC = 0.806 and 0.779, respectively) or the SCC-Ag (AUC = 0.735 and 0.688, respectively) alone in both the training and validation cohorts.

Conclusion

The presented radiomics model can be used for preoperative identification of LNM in patients with early-stage CSCC. Its performance outperforms that of SCC-Ag level analysis alone.

Advances in knowledge

A radiomics model incorporated radiomics signature and SCC-Ag levels demonstrated good performance in identifying LNM in patients with early-stage CSCC.

SUBMITTER: Yan L

PROVIDER: S-EPMC7715994 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:This study aimed to evaluate the feasibility of applying a clinical multimodal radiomics nomogram based on ultrasonography (US) and multiparametric magnetic resonance imaging (MRI) for the prediction of cervical lymph node metastasis (LNM) in papillary thyroid carcinoma (PTC) preoperatively. We performed retrospective evaluations of 133 patients with pathologically confirmed PTC, who were assigned to the training cohort and validation cohort (7 : 3), and extracted radiomics features from the preoperative US, T2-weighted (T2WI),diffusion-weighted (DWI), and contrast-enhanced T1-weighted (CE-T1WI) images. Optimal subsets were selected using minimum redundancy, maximum relevance, and recursive feature elimination in the support vector machine (SVM). For LNM prediction, the radiomics model was constructed by SVM, and Multi-Omics Graph cOnvolutional NETworks (MOGONET) was used for the effective classification of multiradiomics data. Multivariable logistic regression incorporating multiradiomics signatures and clinical risk factors was used to generate a nomogram, whose performance and clinical utility were assessed. Results showed that the nine most predictive features were separately selected from US, T2WI, DWI, and CE-T1WI images, and 18 features were selected in the combined model. The combined radiomics model showed better performance than models based on US, T2WI, DWI, and CE-T1WI. In a comparison of the combined radiomics and MOGONET model, receiver operating curve analysis showed that the area under the curve (AUC) value (95% CI) was 0.84 (0.76-0.93) and 0.84 (0.71-0.96) for the MOGONET model in the training and validation cohorts, respectively. The corresponding values (95% CI) for the combined radiomics model were 0.82 (0.74-0.90) and 0.77 (0.61-0.94), respectively. The MOGONET model had better performance and better prediction specificity compared with the combined radiomics model. The nomogram including the MOGONET signature showed a better predictive value (AUC: 0.81 vs. 0.88) in the training and validation (AUC: 0.74vs. 0.87) cohorts, as compared with the clinical model. Calibration curves showed good agreement in both cohorts. The applicability of the clinical multimodal radiomics (CMR) nomogram in clinical settings was validated by decision curve analysis. In patients with PTC, the CMR nomogram could improve the prediction of cervical LNM preoperatively and may be helpful in clinical decision-making.