Papillary thyroid carcinoma: an ultrasound-based nomogram improves the prediction of lymph node metastases in the central compartment

Deep learning based analysis of dynamic video ultrasonography for predicting cervical lymph node metastasis in papillary thyroid carcinoma

BACKGROUND

Cervical lymph node metastasis (CLNM) is the most common form of thyroid cancer metastasis. Accurate preoperative CLNM diagnosis is of more importance in patients with papillary thyroid cancer (PTC). However, there is currently no unified methods to objectively predict CLNM risk from ultrasonography in PTC patients.This study aimed to develop a deep learning (DL) model to help clinicians more accurately determine the existence of CLNM risk in patients with PTC and then assist them with treatment decisions.

METHODS

Ultrasound dynamic videos of 388 patients with 717 thyroid nodules were retrospectively collected from Zhejiang Cancer Hospital between January 2020 and June 2022. Five deep learning (DL) models were investigated to examine its efficacy for predicting CLNM risks and their performances were also compared with those predicted using two-dimensional ultrasound static images.

RESULTS

In the testing dataset (n = 78), the DenseNet121 model trained on ultrasound dynamic videos outperformed the other four DL models as well as the DL model trained using the two-dimensional (2D) static images across all metrics. Specifically, using DenseNet121, the comparison between the 3D model and 2D model for all metrics are shown as below: AUROC: 0.903 versus 0.828, sensitivity: 0.877 versus 0.871, specificity: 0.865 versus 0.659.

CONCLUSIONS

This study demonstrated that the DenseNet121 model has the greatest potential in distinguishing CLNM from non-CLNM in patients with PTC. Dynamic videos also offered more information about the disease states which have proven to be more efficient and robust in identifying CLNM compared to statis images.

The application of a clinical-multimodal ultrasound radiomics model for predicting cervical lymph node metastasis of thyroid papillary carcinoma

Background

PTC (papillary thyroid cancer) is a lymphotropic malignancy associated with cervical lymph node metastasis (CLNM, including central and lateral LNM), which compromises the effect of treatment and prognosis of patients. Accurate preoperative identification will provide valuable reference information for the formulation of diagnostic and treatment strategies. The aim of this study was to develop and validate a clinical-multimodal ultrasound radiomics model for predicting CLNM of PTC.

Methods

One hundred sixty-four patients with PTC who underwent treatment at our hospital between March 2016 and December 2021 were included in this study. The patients were grouped into a training cohort (n=115) and a validation cohort (n=49). Radiomic features were extracted from the conventional ultrasound (US), contrast-enhanced ultrasound (CEUS) and strain elastography-ultrasound (SE-US) images of patients with PTC. Multivariate logistic regression analysis was used to identify the independent risk factors. FAE software was used for radiomic feature extraction and the construction of different prediction models. The diagnostic performance of each model was evaluated and compared in terms of the area under the curve (AUC), sensitivity, specificity, accuracy, negative predictive value (NPV) and positive predictive value (PPV). RStudio software was used to develop the decision curve and assess the clinical value of the prediction model.

Results

The clinical-multimodal ultrasound radiomics model developed in this study can successfully detect CLNM in PTC patients. A total of 3720 radiomic features (930 features per modality) were extracted from the ROIs of the multimodal images, and 15 representative features were ultimately screened. The combined model showed the best prediction performance in both the training and validation cohorts, with AUCs of 0.957 (95% CI: 0.918-0.987) and 0.932 (95% CI: 0.822-0.984), respectively. Decision curve analysis revealed that the combined model was superior to the other models.

Conclusion

The clinical-multimodal ultrasound radiomics model constructed with multimodal ultrasound radiomic features and clinical risk factors has favorable potential and high diagnostic value for predicting CLNM in PTC patients.

Predicting central lymph node metastasis in papillary thyroid cancer: A nomogram based on clinical, ultrasound and contrast‑enhanced computed tomography characteristics

Central lymph node (CLN) status is considered to be an important risk factor in patients with papillary thyroid carcinoma (PTC). The aim of the present study was to identify risk factors associated with CLN metastasis (CLNM) for patients with PTC based on preoperative clinical, ultrasound (US) and contrast-enhanced computed tomography (CT) characteristics, and establish a prediction model for treatment plans. A total of 786 patients with a confirmed pathological diagnosis of PTC between January 2021 to December 2022 were included in the present retrospective study, with 550 patients included in the training group and 236 patients enrolled in the validation group (ratio of 7:3). Based on the preoperative clinical, US and contrast-enhanced CT features, univariate and multivariate logistic regression analyses were used to determine the independent predictive factors of CLNM, and a personalized nomogram was constructed. Calibration curve, receiver operating characteristic (ROC) curve and decision curve analyses were used to assess discrimination, calibration and clinical application of the prediction model. As a result, 38.9% (306/786) of patients with PTC and CLNM(-) status before surgery had confirmed CLNM using postoperative pathology. In multivariate analysis, a young age (≤45 years), the male sex, no presence of Hashimoto thyroiditis, isthmic location, microcalcification, inhomogeneous enhancement and capsule invasion were independent predictors of CLNM in patients with PTC. The nomogram integrating these 7 factors exhibited strong discrimination in both the training group [Area under the curve (AUC)=0.826] and the validation group (AUC=0.818). Furthermore, the area under the ROC curve for predicting CLNM based on clinical, US and contrast-enhanced CT features was higher than that without contrast-enhanced CT features (AUC=0.818 and AUC=0.712, respectively). In addition, the calibration curve was appropriately fitted and decision curve analysis confirmed the clinical utility of the nomogram. In conclusion, the present study developed a novel nomogram for preoperative prediction of CLNM, which could provide a basis for prophylactic central lymph node dissection in patients with PTC.

PSAMMOMA BODIES IN LYMPH NODES OF THE NECK: POSSIBLE PRECURSOR OF LOCOREGIONAL METASTASES OF PAPILLARY THYROID CARCINOMA

BACKGROUND

Papillary thyroid carcinoma (PTC) is the most common type of well-differentiated thyroid cancer accounting for up to 80% of all thyroid neoplasms. Metastases to the regional lymph nodes (RLN) of the neck are a feature of its biological aggressiveness. The presence of psammoma bodies may be considered a pathomorphological feature of PTC in addition to the papillary structure of tumor and specific nuclear changes. The aim of the study was to evaluate a clinical value of psammoma bodies in the RLN of PTC patients.

MATERIALS AND METHODS

91 patients with PTC who were surgically treated at the Verum Expert Clinic were enrolled in the study. The clinical and pathomorphological data were retrieved from the archival medical records.

RESULTS

According to the results of the clinico-morphological analysis, 51 patients (56%) with PTC had metastases in the RLN of the neck, and 40 (44%) patients had no metastases. Among 51 patients with metastases in the RLN, in 4 patients psammoma bodies in the RLN and tumor tissue were identified. In 3 of these 4 patients, the size of the primary PTC tumor was less than 10 mm, but an aggressive cancer course such as significant number of metastases in the RLN or multifocal growth was found in all these cases.

CONCLUSIONS

The presence of psammoma bodies in RLN and primary PTC tumor could be suggested as a predictor of metastasis to lymph nodes. The detection of point echogenic foci in the lymph nodes by ultrasound at the preoperative stage is a sign of psammoma bodies. This finding can be useful for improving the efficacy in selection of surgical treatment tactics for the optimal neck dissection by planning neck dissection in the presence of such point echogenic foci at the preoperative stage and performing regular check-ups of the patients.

The model for predicting the central lymph node metastasis in cN0 papillary thyroid microcarcinoma with Hashimoto's thyroiditis

Background

The relationship between Hashimoto's thyroiditis (HT) and papillary thyroid microcarcinoma (PTMC) is controversial. These include central lymph node metastasis (CLNM), which affects the prognosis of PTMC patients. This study aimed to establish a predictive model combining ultrasonography and clinicopathological features to accurately evaluate latent CLNM in PTMC patients with HT at the clinical lymph node-negative (cN0) stage.

Methods

In this study, 1102 PTMC patients who received thyroidectomy and central cervical lymph node dissection (CLND) from the First Affiliated Hospital of Shandong First Medical University from January 2021 to December 2022 and the 960th Hospital of PLA from January 2021 to December 2022 were jointly collected. The clinical differences between PTMCs with HT and those without HT were compared. A total of 373 PTMCs with HT in cN0 were randomly divided into a training cohort and a validation cohort. By analyzing and screening the risk factors of CLNM, a nomogram model was established and verified. The predictive performance was measured by the receiver operating characteristic (ROC) curve, calibration curve, and clinical decision curve analysis (DCA).

Results

The ratio of central lymph node metastasis (CLNMR) in PTMCs with HT was 0.0% (0.0%, 15.0%) and 7.7% (0.0%, 40.0%) in the non-HT group (P<0.001). Multivariate logistic regression analysis showed that age, gender, calcification, adjacent to trachea or capsule, and TPOAB were predictors of CLNM in PTMCs with HT. The areas under the curve (AUC) of the prediction models in the training cohort and the validation cohort were 0.835 and 0.825, respectively, which showed good differentiation ability. DCA indicates that the prediction model also has high net benefit and clinical practical value.

Conclusion

This study found that CLN involvement was significantly reduced in PTMC patients with HT, suggesting that different methods should be used to predict CLNM in PTMC patients with HT and without HT, to more accurately assist preoperative clinical evaluation. The actual CLNM situation of PTMCs with HT in cN0 can be accurately predicted by the combination of ultrasonography and clinicopathological features.

A nomogram based on multimodal ultrasound and clinical features for the prediction of central lymph node metastasis in unifocal papillary thyroid carcinoma

OBJECTIVES

To build a predictive model for central lymph node metastasis (CLNM) in unifocal papillary thyroid carcinoma (UPTC) using a combination of clinical features and multimodal ultrasound (MUS).

METHODS

This retrospective study, included 390 UPTC patients who underwent MUS between January 2017 and October 2022 and were divided into a training cohort (n = 300) and a validation cohort (n = 90) based on a cut-off date of June 2022. Independent indicators for constructing the predictive nomogram models were identified using multivariate regression analysis. The diagnostic yield of the 3 predictive models was also assessed using the area under the receiver operating characteristic curve (AUC).

RESULTS

Both clinical factors (age, diameter) and MUS findings (microcalcification, virtual touch imaging score, maximal value of virtual touch tissue imaging and quantification) were significantly associated with the presence of CLNM in the training cohort (all P < .05). A predictive model (MUS + Clin), incorporating both clinical and MUS characteristics, demonstrated favourable diagnostic accuracy in both the training cohort (AUC = 0.80) and the validation cohort (AUC = 0.77). The MUS + Clin model exhibited superior predictive performance in terms of AUCs over the other models (training cohort 0.80 vs 0.72, validation cohort 0.77 vs 0.65, P < .01). In the validation cohort, the MUS + Clin model exhibited higher sensitivity compared to the CLNM model for ultrasound diagnosis (81.2% vs 21.6%, P < .001), while maintaining comparable specificity to the Clin model alone (62.3% vs 47.2%, P = .06). The MUS + Clin model demonstrated good calibration and clinical utility across both cohorts.

CONCLUSION

Our nomogram combining non-invasive features, including MUS and clinical characteristics, could be a reliable preoperative tool to predict CLNM treatment of UPTC.

ADVANCES IN KNOWLEDGE

Our study established a nomogram based on MUS and clinical features for predicting CLNM in UPTC, facilitating informed preoperative clinical management and diagnosis.

Clinical value of dual-energy CT for predicting occult metastasis in central neck lymph nodes of papillary thyroid carcinoma

OBJECTIVES

To predict the probability of occult lymph node metastasis (OLNM) in the central cervical by analyzing the dual-energy computed tomography (DECT) parameters derived from papillary thyroid carcinoma (PTC).

METHODS

Data were retrospectively collected from patients with pathologically confirmed PTC who underwent arterial and venous phases of enhanced DECT with concurrent central neck lymph node dissection (CLND). Three clinical features, three shape-related features, and twenty-six DECT-derived parameters were measured. The univariate and multivariate analyses were applied to select the relevant parameters and develop the nomogram.

RESULTS

A total 140 cases with negative diagnosis of cervical central lymph node metastases by preoperative evaluation were included, among which 88 patients with metastasis (OLNM +) and 52 patients without metastasis (OLNM -) were finally confirmed by pathology. (1) Anteroposterior/transverse diameter ratio (A/T) derived from the PTC focus had significant difference between the OLNM + and OLNM - groups (p < 0.05). (2) In the arterial phase, iodine concentration (ICarterial), normalized iodine concentration (NICarterial), effective atomic number (Zeff-arterial), electron density (EDarterial), and slope of energy curve (karterial) from PTC focus showed significant difference (all p < 0.05) between the two groups. In the venous phase, only the CT value under the 40 keV (HU40keVvenous) had differences (p < 0.05). (3) The nomogram was produced to predict the probability of OLNM, and the AUC, sensitivity, and specificity in the training and test cohort were 0.830, 75.0%, 76.9%, and 0.829, 65.9%, 84.6%, respectively.

CONCLUSIONS

DECT parameters combined with shape-related feature derived from PTC might be used as predictors of OLNM in the central neck.

CLINICAL RELEVANCE STATEMENT

Preoperative imaging evaluation combining shape-related features and dual-energy CT parameters could serve as a reference to discern occult lymph node metastasis in central neck during the surgically planning of papillary thyroid carcinoma.

KEY POINTS

• Papillary thyroid carcinoma (PTC) patients may have occult lymph node metastasis (OLNM) in the central neck, which is extremely difficult to find by preoperative imaging examination. • Dual-energy CT quantitative evaluation has higher accuracy than conventional CT and can predicting OLNM in the central neck of PTC. • Dual-energy CT quantitative parameters and morphology of PTC can serve as a useful tool in predicting OLNM in the central neck, and as a guide for personalized treatment.

The diagnostic value of ultrasound on different-sized thyroid nodules based on ACR TI-RADS

OBJECTIVES

The thyroid nodule is one of the most common endocrine system diseases. Risk classification models based on ultrasonic features have been created by multiple professional societies, including the American College of Radiology (ACR), which published the Thyroid Imaging Reporting and Data System (TI-RADS) in 2017. The effect of the size in the diagnostic value of ultrasound remains not well defined. The purposes of our study aims to explore diagnostic value of the ACR TI-RADS on different-sized thyroid nodules.

METHODS

A total of 1183 thyroid nodules were selected from 952 patients with thyroid nodules confirmed by surgical pathology from January 2021 to October 2022. Based on the maximum diameters of the nodules, they were stratified into groups A ( ≤ 10 mm), B ( > 10 mm, < 20 mm) and C ( ≥ 20 mm). The ultrasonic features of the thyroid nodules in each group were evaluated and scored based on ACR TI-RADS, and the receiver operating characteristic curve (ROC) was plotted to determine the optimal cut-off value for the ACR TI-RADS scores and categories in each group. Finally, the diagnostic efficacy of ACR TI-RADS on different-sized thyroid nodules was analyzed.

RESULTS

Among the 1183 thyroid nodules, 340 were benign, 10 were low-risk and 833 were malignant. For the convenience of statistical analysis, low-risk thyroid nodules were classified as malignant in this study. The ACR TI-RADS scores and categorical levels of malignant thyroid nodules in each group were higher than those of benign ones (p < 0.05). The areas under the ROCs (AUCs) plotted based on scores were 0.741, 0.907, and 0.904 respectively in the three groups, and the corresponding optimal cut-off values were > 6 points, > 5 points and > 4 points respectively. While the AUCs of the ACR TI-RADS categories were 0.668, 0.855, and 0.887 respectively in each group, with the optimal cut-off values were all > TR4. Besides, for thyroid nodules of larger sizes, ACR TI-RADS exhibited weaker sensitivity with lower positive prediction value (PPV), but the specificity and negative prediction value (NPV) were both higher, presenting with statistically significant differences (p < 0.05).

CONCLUSION

For thyroid nodules of different sizes, the diagnostic efficacy of ACR TI-RADS varies as well. The system shows better diagnostic efficacy on thyroid nodules of > 10 mm than on those ≤ 10 mm. Considering the favorable prognosis of thyroid microcarcinoma and the low diagnostic efficacy of ACR TI-RADS on it, the scoring and classification of thyroid micro-nodules can be left out in appropriate cases, so as to avoid the over-diagnosis and over-treatment of thyroid microcarcinoma to a certain extent.

Ultrasound radiomics models based on multimodal imaging feature fusion of papillary thyroid carcinoma for predicting central lymph node metastasis

OBJECTIVE

This retrospective study aimed to establish ultrasound radiomics models to predict central lymph node metastasis (CLNM) based on preoperative multimodal ultrasound imaging features fusion of primary papillary thyroid carcinoma (PTC).

METHODS

In total, 498 cases of unifocal PTC were randomly divided into two sets which comprised 348 cases (training set) and 150 cases (validition set). In addition, the testing set contained 120 cases of PTC at different times. Post-operative histopathology was the gold standard for CLNM. The following steps were used to build models: the regions of interest were segmented in PTC ultrasound images, multimodal ultrasound image features were then extracted by the deep learning residual neural network with 50-layer network, followed by feature selection and fusion; subsequently, classification was performed using three classical classifiers-adaptive boosting (AB), linear discriminant analysis (LDA), and support vector machine (SVM). The performances of the unimodal models (Unimodal-AB, Unimodal-LDA, and Unimodal-SVM) and the multimodal models (Multimodal-AB, Multimodal-LDA, and Multimodal-SVM) were evaluated and compared.

RESULTS

The Multimodal-SVM model achieved the best predictive performance than the other models (P < 0.05). For the Multimodal-SVM model validation and testing sets, the areas under the receiver operating characteristic curves (AUCs) were 0.910 (95% CI, 0.894-0.926) and 0.851 (95% CI, 0.833-0.869), respectively. The AUCs of the Multimodal-SVM model were 0.920 (95% CI, 0.881-0.959) in the cN0 subgroup-1 cases and 0.828 (95% CI, 0.769-0.887) in the cN0 subgroup-2 cases.

CONCLUSION

The ultrasound radiomics model only based on the PTC multimodal ultrasound image have high clinical value in predicting CLNM and can provide a reference for treatment decisions.

Development and validation of a dynamic nomogram based on conventional ultrasound and contrast-enhanced ultrasound for stratifying the risk of central lymph node metastasis in papillary thyroid carcinoma preoperatively

Purpose

The aim of this study was to develop and validate a dynamic nomogram by combining conventional ultrasound (US) and contrast-enhanced US (CEUS) to preoperatively evaluate the probability of central lymph node metastases (CLNMs) for patients with papillary thyroid carcinoma (PTC).

Methods

A total of 216 patients with PTC confirmed pathologically were included in this retrospective and prospective study, and they were divided into the training and validation cohorts, respectively. Each cohort was divided into the CLNM (+) and CLNM (-) groups. The least absolute shrinkage and selection operator (LASSO) regression method was applied to select the most useful predictive features for CLNM in the training cohort, and these features were incorporated into a multivariate logistic regression analysis to develop the nomogram. The nomogram's discrimination, calibration, and clinical usefulness were assessed in the training and validation cohorts.

Results

In the training and validation cohorts, the dynamic nomogram (https://clnmpredictionmodel.shinyapps.io/PTCCLNM/) had an area under the receiver operator characteristic curve (AUC) of 0.844 (95% CI, 0.755-0.905) and 0.827 (95% CI, 0.747-0.906), respectively. The Hosmer-Lemeshow test and calibration curve showed that the nomogram had good calibration (p = 0.385, p = 0.285). Decision curve analysis (DCA) showed that the nomogram has more predictive value of CLNM than US or CEUS features alone in a wide range of high-risk threshold. A Nomo-score of 0.428 as the cutoff value had a good performance to stratify high-risk and low-risk groups.

Conclusion

A dynamic nomogram combining US and CEUS features can be applied to risk stratification of CLNM in patients with PTC in clinical practice.

Nomogram for preoperative estimation risk of lateral cervical lymph node metastasis in papillary thyroid carcinoma: a multicenter study

BACKGROUND

Lateral lymph node metastasis (LLNM) is frequent in papillary thyroid carcinoma (PTC) and is associated with a poor prognosis. This study aimed to developed a clinical-ultrasound (Clin-US) nomogram to predict LLNM in patients with PTC.

METHODS

In total, 2612 PTC patients from two hospitals (H1: 1732 patients in the training cohort and 578 patients in the internal testing cohort; H2: 302 patients in the external testing cohort) were retrospectively enrolled. The associations between LLNM and preoperative clinical and sonographic characteristics were evaluated by the univariable and multivariable logistic regression analysis. The Clin-US nomogram was built basing on multivariate logistic regression analysis. The predicting performance of Clin-US nomogram was evaluated by calibration, discrimination and clinical usefulness.

RESULTS

The age, gender, maximum diameter of tumor (tumor size), tumor position, internal echo, microcalcification, vascularization, mulifocality, and ratio of abutment/perimeter (A/P) > 0.25 were independently associated with LLNM metastatic status. In the multivariate analysis, gender, tumor size, mulifocality, position, microcacification, and A/P > 0.25 were independent correlative factors. Comparing the Clin-US nomogram and US features, Clin-US nomogram had the highest AUC both in the training cohort and testing cohorts. The Clin‑US model revealed good discrimination between PTC with LLNM and without LLNM in the training cohort (AUC = 0.813), internal testing cohort (AUC = 0.815) and external testing cohort (AUC = 0.870).

CONCLUSION

Our findings suggest that the ClinUS nomogram we newly developed can effectively predict LLNM in PTC patients and could help clinicians choose appropriate surgical procedures.

Nomogram incorporating preoperative clinical and ultrasound indicators to predict aggressiveness of solitary papillary thyroid carcinoma

Objective

To construct a nomogram based on preoperative clinical and ultrasound indicators to predict aggressiveness of solitary papillary thyroid carcinoma (PTC).

Methods

Preoperative clinical and ultrasound data from 709 patients diagnosed with solitary PTC between January 2017 and December 2020 were analyzed retrospectively. Univariate and multivariate logistic regression analyses were performed to identify the factors associated with PTC aggressiveness, and these factors were used to construct a predictive nomogram. The nomogram's performance was evaluated in the primary and validation cohorts.

Results

The 709 patients were separated into a primary cohort (n = 424) and a validation cohort (n = 285). Univariate analysis in the primary cohort showed 13 variables to be associated with aggressive PTC. In multivariate logistic regression analysis, the independent predictors of aggressive behavior were age (OR, 2.08; 95% CI, 1.30-3.35), tumor size (OR, 4.0; 95% CI, 2.17-7.37), capsule abutment (OR, 2.53; 95% CI, 1.50-4.26), and suspected cervical lymph nodes metastasis (OR, 2.50; 95% CI, 1.20-5.21). The nomogram incorporating these four predictors showed good discrimination and calibration in both the primary cohort (area under the curve, 0.77; 95% CI, 0.72-0.81; Hosmer-Lemeshow test, P = 0.967 and the validation cohort (area under the curve, 0.72; 95% CI, 0.66-0.78; Hosmer-Lemeshow test, P = 0.251).

Conclusion

The proposed nomogram shows good ability to predict PTC aggressiveness and could be useful during treatment decision making.

Advances in knowledge

Our nomogram-based on four indicators-provides comprehensive assessment of aggressive behavior of PTC and could be a useful tool in the clinic.

Dual-modal radiomics for predicting cervical lymph node metastasis in papillary thyroid carcinoma

BACKGROUND

Preoperative prediction of cervical lymph node metastasis (CLNM) in patients with papillary thyroid carcinoma (PTC) is significant for surgical decision-making.

OBJECTIVE

This study aims to develop a dual-modal radiomics (DMR) model based on grayscale ultrasound (GSUS) and dual-energy computed tomography (DECT) for non-invasive CLNM in PTC.

METHODS

In this study, 348 patients with pathologically confirmed PTC at Jiangsu University Affiliated People's Hospital who completed preoperative ultrasound (US) and DECT examinations were enrolled and randomly assigned to training (n = 261) and test (n = 87) cohorts. The enrolled patients were divided into two groups based on pathology findings namely, CLNM (n = 179) and CLNM-Free (n = 169). Radiomics features were extracted from GSUS images (464 features) and DECT images (960 features), respectively. Pearson correlation coefficient (PCC) and the least absolute shrinkage and selection operator (LASSO) regression with 10-fold cross-validation were then used to select CLNM-related features. Based on the selected features, GSUS, DECT, and GSUS combined DECT radiomics models were constructed by using a Support Vector Machine (SVM) classifier.

RESULTS

Three predictive models based on GSUS, DECT, and a combination of GSUS and DECT, yielded performance of areas under the curve (AUC) = 0.700 [95% confidence interval (CI), 0.662-0.706], 0.721 [95% CI, 0.683-0.727], and 0.760 [95% CI, 0.728-0.762] in the training dataset, and AUC = 0.643 [95% CI, 0.582-0.734], 0.680 [95% CI, 0.623-0.772], and 0.744 [95% CI, 0.686-0.784] in the test dataset, respectively. It shows that the predictive model combined GSUS and DECT outperforms both models using GSUS and DECT only.

CONCLUSIONS

The newly developed combined radiomics model could more accurately predict CLNM in PTC patients and aid in better surgical planning.

Radiomics Analysis of Gray-Scale Ultrasonographic Images of Papillary Thyroid Carcinoma > 1 cm: Potential Biomarker for the Prediction of Lymph Node Metastasis

Purpose

This study aimed to investigate radiomics analysis of ultrasonographic images to develop a potential biomarker for predicting lymph node metastasis in papillary thyroid carcinoma (PTC) patients.

Materials and Methods

This study included 431 PTC patients from August 2013 to May 2014 and classified them into the training and validation sets. A total of 730 radiomics features, including texture matrices of gray-level co-occurrence matrix and gray-level run-length matrix and single-level discrete two-dimensional wavelet transform and other functions, were obtained. The least absolute shrinkage and selection operator method was used for selecting the most predictive features in the training data set.

Results

Lymph node metastasis was associated with the radiomics score (p < 0.001). It was also associated with other clinical variables such as young age (p = 0.007) and large tumor size (p = 0.007). The area under the receiver operating characteristic curve was 0.687 (95% confidence interval: 0.616-0.759) for the training set and 0.650 (95% confidence interval: 0.575-0.726) for the validation set.

Conclusion

This study showed the potential of ultrasonography-based radiomics to predict cervical lymph node metastasis in patients with PTC; thus, ultrasonography-based radiomics can act as a biomarker for PTC.

Extreme gradient boosting model to assess risk of central cervical lymph node metastasis in patients with papillary thyroid carcinoma: Individual prediction using SHapley Additive exPlanations

BACKGROUND AND OBJECTIVES

Central cervical lymph node metastasis (CLNM) is considered a risk factor for recurrence in patients with papillary thyroid carcinoma (PTC). Traditional machine learning models suffered from "black-box" problems, which could not exactly explain the interactive effects of the risk factors. We aimed to develop an eXtreme Gradient Boosting (XGBoost) model to assess CLNM, including positive and negative effects.

METHODS

1,122 patients with PTC admitted at Tianjin First Central Hospital from 2016 to 2020 were retrospectively selected. They were randomly divided into the training and test datasets with an 8:2 ratio. 108 patients with PTC admitted at Binzhou Medical University Hospital in 2020 served as the validation dataset. The XGBoost model was used to assess CLNM. The 10-fold cross-validation was utilized for model selection, and the metric used to evaluate classification performance was the average area under the curve (AUC) of 10-fold cross-validation. Interpretation and transparency of the "black-box" problem were performed. SHapley Additive exPlanations (SHAP) and local interpretable model-agnostic explanation (LIME) were used to ensure the stability and reliability of the model.

RESULTS

The XGBoost model based on ultrasound and dual-energy computed tomography images of the solitary primary lesion had an excellent performance for assessing CLNM, with average AUCs of 0.918, 0.903, and 0.881 in the training, test, and validation datasets, respectively. SHAP plots showed the influence of each parameter on the XGBoost model, including positive (i.e., capsular invasion, diameter, iodine concentration in the venous phase, and calcification) and negative (i.e., sex and age) impacts. For all cases, the capsular invasion prediction weight was the highest; for individual cases, different predictors were assigned different weights. Moreover, the performance of the XGBoost model was better than classical machine-learning models.

CONCLUSIONS

This study developed and validated an XGBoost model for assessing CLNM in patients with PTC. The ability to visually interpret the positive and negative effects made the XGBoost model an effective tool for guiding clinical treatment.

Development and validation of an individualized nomogram for predicting the high-volume (> 5) central lymph node metastasis in papillary thyroid microcarcinoma

PURPOSE

Papillary thyroid microcarcinoma (PTMC) frequently presents a favorable clinical outcome, while aggressive invasiveness can also be found in some of this population. Identifying the risk clinical factors of high-volume (> 5) central lymph node metastasis (CLNM) in PTMC patients could help oncologists make a better-individualized clinical decision.

METHODS

We retrospectively reviewed the clinical characteristics of adult patients with PTC in the Surveillance, Epidemiology, and End Results (SEER) database between Jan 2010 and Dec 2015 and in one medical center affiliated to Chongqing Medical University between Jan 2018 and Oct 2020. Univariate and multivariate logistic regression analyses were used to determine the risk factors for high volume of CLNM in PTMC patients.

RESULTS

The male gender (OR = 2.02, 95% CI 1.46-2.81), larger tumor size (> 5 mm, OR = 1.64, 95% CI 1.13-2.38), multifocality (OR = 1.87, 95% CI 1.40-2.51), and extrathyroidal invasion (OR = 3.67; 95% CI 2.64-5.10) were independent risk factors in promoting high-volume of CLNM in PTMC patients. By contrast, elderly age (≥ 55 years) at diagnosis (OR = 0.57, 95% CI 0.40-0.81) and PTMC-follicular variate (OR = 0.60, 95% CI 0.42-0.87) were determined as the protective factors. Based on these indicators, a nomogram was further constructed with a good concordance index (C-index) of 0.702, supported by an external validating cohort with a promising C-index of 0.811.

CONCLUSION

A nomogram was successfully established and validated with six clinical indicators. This model could help surgeons to make a better-individualized clinical decision on the management of PTMC patients, especially in terms of whether prophylactic central lymph node dissection and postoperative radiotherapy should be warranted.

Nomograms based on preoperative multimodal ultrasound of papillary thyroid carcinoma for predicting central lymph node metastasis

OBJECTIVES

To establish a nomogram for predicting central lymph node metastasis (CLNM) based on the preoperative clinical and multimodal ultrasound (US) features of papillary thyroid carcinoma (PTC) and cervical LNs.

METHODS

Overall, 822 patients with PTC were included in this retrospective study. A thyroid tumor ultrasound model (TTUM) and thyroid tumor and cervical LN ultrasound model (TTCLNUM) were constructed as nomograms to predict the CLNM risk. Areas under the curve (AUCs) evaluated model performance. Calibration and decision curves were applied to assess the accuracy and clinical utility.

RESULTS

For the TTUM training and test sets, the AUCs were 0.786 and 0.789 and bias-corrected AUCs were 0.786 and 0.831, respectively. For the TTCLNUM training and test sets, the AUCs were 0.806 and 0.804 and bias-corrected AUCs were 0.807 and 0.827, respectively. Calibration and decision curves for the TTCLNUM nomogram exhibited higher accuracy and clinical practicability. The AUCs were 0.746 and 0.719 and specificities were 0.942 and 0.905 for the training and test sets, respectively, when the US tumor size was ≤ 8.45 mm, while the AUCs were 0.737 and 0.824 and sensitivity were 0.905 and 0.880, respectively, when the US tumor size was > 8.45 mm.

CONCLUSION

The TTCLNUM nomogram exhibited better predictive performance, especially for the CLNM risk of different PTC tumor sizes. Thus, it serves as a useful clinical tool to supply valuable information for active surveillance and treatment decisions.

KEY POINTS

• Our preoperative noninvasive and intuitive prediction method can improve the accuracy of central lymph node metastasis (CLNM) risk assessment and guide clinical treatment in line with current trends toward personalized treatments. • Preoperative clinical and multimodal ultrasound features of primary papillary thyroid carcinoma (PTC) tumors and cervical LNs were directly used to build an accurate and easy-to-use nomogram for predicting CLNM. • The thyroid tumor and cervical lymph node ultrasound model exhibited better performance for predicting the CLNM of different PTC tumor sizes. It may serve as a useful clinical tool to provide valuable information for active surveillance and treatment decisions.

Real-Time Elastography: A Web-Based Nomogram Improves the Preoperative Prediction of Central Lymph Node Metastasis in cN0 PTC

BACKGROUND

Given the difficulty of accurately determining the central lymph node metastasis (CLNM) status of patients with clinically node-negative (cN0) papillary thyroid carcinoma (PTC) before surgery, this study aims to combine real-time elastography (RTE) and conventional ultrasound (US) features with clinical features. The information is combined to construct and verify the nomogram to foresee the risk of CLNM in patients with cN0 PTC and to develop a network-based nomogram.

METHODS

From January 2018 to February 2020, 1,157 consecutive cases of cN0 PTC after thyroidectomy and central compartment neck dissection were retrospectively analyzed. The patients were indiscriminately allocated (2:1) to a training cohort (771 patients) and validation cohort (386 patients). Multivariate logistic regression analysis of US characteristics and clinical information in the training cohort was performed to screen for CLNM risk predictors. RTE data were included to construct prediction model 1 but were excluded when constructing model 2. DeLong's test was used to select a forecast model with better receiver operator characteristic curve performance to establish a web-based nomogram. The clinical applicability, discrimination, and calibration of the preferable prediction model were assessed.

RESULTS

Multivariate regression analysis showed that age, sex, tumor size, bilateral tumors, the number of tumor contacting surfaces, chronic lymphocytic thyroiditis, and RTE were risk predictors of CLNM in cN0 PTC patients, which constituted prediction model 1. Model 2 included the first six risk predictors. Comparison of the areas under the curves of the two models showed that model 1 had better prediction performance (training set 0.798 vs. 0.733, validation set 0.792 vs. 0.715, p < 0.001) and good discrimination and calibration. RTE contributed significantly to the performance of the prediction model. Decision curve analysis showed that patients could obtain good net benefits with the application of model 1.

CONCLUSION

A noninvasive web-based nomogram combining US characteristics and clinical risk factors was developed in the research. RTE could improve the prediction accuracy of the model. The dynamic nomogram has good performance in predicting the probability of CLNM in cN0 PTC patients.

Establishing a Predictive Nomogram for Cervical Lymph Node Metastasis in Patients With Papillary Thyroid Carcinoma

Objectives

The purpose of this study was to establish a nomogram for predicting cervical lymph node metastasis (CLNM) in patients with papillary thyroid carcinoma (PTC).

Materials and Methods

A total of 418 patients with papillary thyroid carcinoma undergoing total thyroidectomy with cervical lymph node dissection were enrolled in the retrospective study from January 2016 to September 2019. Univariate and multivariate Logistic regression analysis were performed to screen the clinicopathologic, laboratory and ultrasound (US) parameters influencing cervical lymph nodes metastasis and develop the predicting model.

Results

CLNM was proved in 34.4% (144/418) of patients. In the multivariate regression analysis, Male, Age < 45 years, Tumor size > 20mm, multifocality, ambiguous boundary, extracapsular invasion and US-suggested lymph nodes metastasis were independent risk factors of CLNM (p < 0.05). Prediction nomogram showed an excellent discriminative ability, with a C-index of 0.940 (95% confidence interval [CI], 0.888-0.991), and a good calibration.

Conclusion

The established nomogram showed a good prediction of CLNM in patients with PTC. It is conveniently used and should be considered in the determination of surgical procedures.

A Clinical Predictive Model of Central Lymph Node Metastases in Papillary Thyroid Carcinoma

BACKGROUND

Thyroid carcinoma is one of the most common endocrine tumors, and papillary thyroid carcinoma (PTC) is the most common pathological type. Current studies have reported that PTC has a strong propensity for central lymph node metastases (CLNMs). Whether to prophylactically dissect the central lymph nodes in PTC remains controversial. This study aimed to explore the risk factors and develop a predictive model of CLNM in PTC.

METHODS

A total of 2,554 patients were enrolled in this study. The basic information, laboratory examination, characteristics of cervical ultrasound, genetic test, and pathological diagnosis were collected. The collected data were analyzed by univariate logistic analysis and multivariate logistic analysis. The risk factors were evaluated, and the predictive model was constructed of CLNM.

RESULTS

The multivariate logistic analysis showed that Age (p < 0.001), Gender (p < 0.001), Multifocality (p < 0.001), BRAF (p = 0.027), and Tumor size (p < 0.001) were associated with CLNM. The receiver operating characteristic curve (ROC curve) showed high efficiency with an area under the ROC (AUC) of 0.781 in the training group. The calibration curve and the calibration of the model were evaluated. The decision curve analysis (DCA) for the nomogram showed that the nomogram can provide benefits in this study.

CONCLUSION

The predictive model of CLNM constructed and visualized based on the evaluated risk factors was confirmed to be a practical and convenient tool for clinicians to predict the CLNM in PTC.

Hashimoto's Thyroiditis Is Associated With Central Lymph Node Metastasis in Classical Papillary Thyroid Cancer: Analysis from a High-Volume Single-Center Experience

PURPOSE

Central lymph node metastasis (CLNM) is regarded as a predictor for local recurrence in patients with papillary thyroid carcinoma (PTC) but the role of prophylactic central lymph node dissection (CLND) is controversial. Our study aims to identify the clinical factors associated with CLNM and develop a nomogram for making individualized clinical decisions.

METHOD

The perioperative data of 1,054 consecutive patients between Jan 2019 and April 2021, in our center, were reviewed and analyzed. A total of 747 patients with histopathologically confirmed classical PTC were included as the training cohort and 374 (50% training cases) patients were randomly selected to build a validating cohort via internal bootstrap analysis. Univariate and multivariate logistic regression were used to analyze the correlation between clinicopathological characteristics and CLNM.

RESULT

In the training cohort, 33.6% (251/747) of patients with classical PTC were confirmed with CLNM. And the CLNM was determined in 31.4% (168/535) of non-Hashimoto's thyroiditis (HT) patients versus 39.2% (83/212) in HT patients (p=0.043). Four factors including gender, age, size, and HT status were confirmed significantly associated with CLNM. The established nomogram showed good discrimination and consistency with a C-index of 0.703, supported by the internal validation cohort with a C-index of 0.701. The decision curve analysis showed the nomogram has promising clinical feasibility.

CONCLUSION

Our study suggested that classical PTC patients with features like male gender, age<55 years old, tumor size>1cm, and HT condition had a higher risk of CLNM. And the nomogram we developed can help surgeons make individualized clinical decisions in classical PTC patients during preoperative and intraoperative management.

A radiomics nomogram for the ultrasound-based evaluation of central cervical lymph node metastasis in papillary thyroid carcinoma

PURPOSE

To develop and validate a radiomics nomogram based on ultrasound (US) to predict central cervical lymph node (LN) metastasis in patients with papillary thyroid carcinoma (PTC).

METHODS

PTC patients with pathologically confirmed presence or absence of central cervical LN metastasis in our hospital between March 2021 and November 2021 were enrolled as the training cohort. Radiomics features were extracted from the preoperative US images, and a radiomics signature was constructed. Univariate and multivariate logistic regression analyses were used to screen out the independent risk factors, and a radiomics nomogram was established. The performance of the model was verified in the independent test cohort of PTC patients who underwent thyroidectomy and cervical LN dissection in our hospital from December 2021 to March 2022.

RESULTS

In the independent test cohort, the radiomics model based on long-axis cross-section and short-axis cross-section images outperformed the radiomics models based on either one of these sections (the area under the curve (AUC), 0.69 vs. 0.62 and 0.66). The radiomics signature consisted of 4 selected features. The US radiomics nomogram included the radiomics signature, age, gender, BRAF V600E mutation status, and extrathyroidal extension (ETE) status. In the independent test cohort, the AUC of the receiver operating curve(ROC) of this nomogram was 0.76, outperformingthe clinical model and the radiomics model (0.63 and 0.69, respectively), and also much better than preoperative US examination (AUC, 0.60). Decision curve analysis indicated that the radiomics nomogram was clinically useful.

CONCLUSIONS

This study presents an efficient and useful US radiomics nomogram that can provide comprehensive information to assist clinicians in the individualized preoperative prediction of central cervical LN metastasis in PTC patients.

Image Segmentation of Thyroid Nodule and Capsule for Diagnosing Central Compartment Lymph Node Metastasis

Thyroid ultrasound (US) image segmentation is of great significance for both doctors and patients. However, it is a challenging task because of the low image quality, low contrast and complex background in each US image. In recent years, some researchers have done thyroid nodule segmentation tasks, but the results achieved are not particularly satisfactory. In this paper, we have broadened the targets of interest and included both thyroid nodules and capsules into our research scope. We propose a method that implements a C-MMDetection to detect and extract the region of interest (ROI), and a modified salient object detection network U²-RNet to segment nodules and capsules respectively. Experiments show that our method segments nodules and capsules in US images more effectively than other networks, which is very helpful for doctors to diagnose central compartment lymph node metastasis (CLNM).

Construction and validation of a nomogram for predicting cervical lymph node metastasis in classic papillary thyroid carcinoma

PURPOSE

Patients with papillary thyroid carcinoma (PTC) frequently present a relatively poor prognosis when they coexist with cervical lymph node metastasis (LNM). Moreover, it remains controversial whether prophylactic lymph node dissection (LND) should be performed for patients without clinically lymph node metastasis. Thus, we hereby develop a nomogram for predicting the cervical LNM (including central and lateral LNM) in patients with PTC.

METHODS

We retrospectively reviewed the clinical characteristics of adult patients with PTC in the surveillance, epidemiology, and end results (SEER) database between 2010 and 2015 and in our Department of Breast and Thyroid Surgery in the Second Affiliated Hospital of Chongqing Medical University between 2019 and 2020.

RESULT

A total of 21,972 patients in the SEER database and 747 patients in our department who met the inclusion criteria were enrolled in this study. Ultimately, six clinical features including age, gender, race, extrathyroidal invasion, multifocality, and tumor size were identified to be associated with cervical LNM in patients with PTC, which were screened to develop a nomogram. This model had satisfied discrimination with a concordance index (C-index) of 0.733, supported by both internal and external validation with a C-index of 0.731 and 0.716, respectively. A decision curve analysis was subsequently made to evaluate the feasibility of this nomogram for predicting cervical LNM. Besides, a positive correlation between nomogram score and the average number of lymph node metastases was observed in all groups.

CONCLUSION

This visualized multipopulational-based nomogram model was successfully established. We determined that various clinical characteristics were significantly associated with cervical LNM, which would be better helping clinicians make individualized clinical decisions for PTC patients.

Nomogram to assess risk of central cervical lymph node metastasis in patients with cN0 papillary thyroid carcinoma

OBJECTIVE

This study developed and validated an individualized prediction risk model for the need for central cervical lymph node dissection in patients with cN0 papillary thyroid carcinoma (PTC) diagnosed by ultrasound.

METHODS

Upon retrospective review, derivation and internal validation cohorts comprised 1585 consecutive patients with PTC treated from January 2017 to December 2019 at Hospital A. The external validation cohort consisted of 406 consecutive patients treated at Hospital B from January 2016 to June 2020. Independent risk factors for central cervical lymph node metastasis (CLNM) were determined through univariable and multivariable logistic regression analysis. An individualized risk prediction model was constructed and illustrated as a nomogram, which was internally and externally validated.

RESULTS

The following risk factors of CLNM were established: the solitary primary thyroid nodule's diameter, shape, calcification, and capsular abutment-to-lesion perimeter ratio. The areas under the receiver operating characteristic curves of the risk prediction model for the internal and external validation cohorts were 0.921 and 0.923, respectively. The calibration curve showed good agreement between the nomogram-estimated probability of CLNM and the actual CLNM rate in the three cohorts. The decision curve analysis confirmed the clinical usefulness of the nomogram.

CONCLUSION

This study developed and validated a model for predicting risk of CLNM in the individual patient with cN0 PTC, which should be an efficient tool for guiding clinical treatment.

Ultrasound lymphatic imaging for the diagnosis of metastatic central lymph nodes in papillary thyroid cancer

OBJECTIVES

Up to 40% of papillary thyroid cancer (PTC) patients have lymph node metastasis, a condition that implies persistent, recurrent, or progressive disease. However, the American Joint Committee on Cancer Manual states that there is no reliable examination for adequate lymph node staging. Therefore, our aim is to develop a lymphatic imaging technique using ultrasonography to address this challenge.

METHODS

We consecutively enrolled PTC patients who underwent ultrasound (US) lymphatic imaging via the peritumoral injection of contrast media. Identification of the sentinel lymph nodes and the targeted sentinel lymph nodes was separately based on the lymphatic drainage pathway and the enhancement patterns. Every identified targeted node was assigned a score, according to the features on conventional US and enhancement patterns, and was referred for ultrasound-guided fine-needle aspiration. Cytological and histopathologic results represented the statuses of the targeted lymph nodes and overall central lymph nodes, respectively, which were applied to evaluate the diagnostic performance of US lymphatic imaging.

RESULTS

In total, 100 PTC patients were included. On the basis of the cytological results, the sensitivity (97.1%, 95% confidence interval [CI]: 84.7-99.9%) of detecting positive targeted nodes by US lymphatic imaging significantly increased by 45.5% at a threshold of 4 or higher (p = 0.0001), without loss of specificity (p = 1.0000). The surgical results showed that the metastatic degree was positively correlated with an increase in the score (τ: 0.671, p < 0.001).

CONCLUSION

Ultrasound lymphatic imaging has a high diagnostic performance, and its corresponding scoring system facilitates grading of the nodal burden in the central compartment.

KEY POINTS

• Ultrasound neck lymphatic imaging is an effective contrast-enhanced ultrasound (CEUS) technique (applied after the peritumoral injection of contrast media) for identifying sentinel lymph nodes in the central compartment by tracing the imaged afferent lymphatic vessel. • Lack of enhancement or perfusion defects is the typical enhancement pattern for recognizing the involved central lymph nodes. • Ultrasound lymphatic imaging for identification of positive central lymph nodes before surgery may effectively avoid complications associated with the surgical sentinel node procedure.

A Radiomic Nomogram for the Ultrasound-Based Evaluation of Extrathyroidal Extension in Papillary Thyroid Carcinoma

Purpose

To construct a sequence diagram based on radiological and clinical factors for the evaluation of extrathyroidal extension (ETE) in patients with papillary thyroid carcinoma (PTC).

Materials and Methods

Between January 2016 and January 2020, 161 patients with PTC who underwent preoperative ultrasound examination in the Affiliated People’s Hospital of Jiangsu University were enrolled in this retrospective study. According to the pathology results, the enrolled patients were divided into a non-ETE group and an ETE group. All patients were randomly divided into a training cohort (n = 97) and a validation cohort (n = 64). A total of 479 image features of lesion areas in ultrasonic images were extracted. The radiomic signature was developed using least absolute shrinkage and selection operator algorithms after feature selection using the minimum redundancy maximum relevance method. The radiomic nomogram model was established by multivariable logistic regression analysis based on the radiomic signature and clinical risk factors. The discrimination, calibration, and clinical usefulness of the nomogram model were evaluated in the training and validation cohorts.

Results

The radiomic signature consisted of six radiomic features determined in ultrasound images. The radiomic nomogram included the parameters tumor location, radiological ETE diagnosis, and the radiomic signature. Area under the curve (AUC) values confirmed good discrimination of this nomogram in the training cohort [AUC, 0.837; 95% confidence interval (CI), 0.756–0.919] and the validation cohort (AUC, 0.824; 95% CI, 0.723–0.925). The decision curve analysis showed that the radiomic nomogram has good clinical application value.

Conclusion

The newly developed radiomic nomogram model is a noninvasive and reliable tool with high accuracy to predict ETE in patients with PTC.

Nomogram for Preoperative Estimation of Cervical Lymph Node Metastasis Risk in Papillary Thyroid Microcarcinoma

OBJECTIVE

Accurate preoperative identification of cervical lymph node metastasis (CLNM) is essential for clinical management and established of different surgical protocol for patients with papillary thyroid microcarcinoma (PTMC). Herein, we aimed to develop an ultrasound (US) features and clinical characteristics-based nomogram for preoperative diagnosis of CLNM for PTMC.

METHOD

Our study included 552 patients who were pathologically diagnosed with PTMC between January 2015 and June 2019. All patients underwent total thyroidectomy or lobectomy and divided into two groups: CLNM and non-CLNM. Univariate and multivariate analysis were performed to examine risk factors associated with CLNM. A nomogram comprising the prognostic model to predict the CLNM was established, and internal validation in the cohort was performed.

RESULTS

CLNM and non-CLNM were observed in 216(39.1%) and 336(60.9%) cases, respectively. Seven variables of clinical and US features as potential predictors including male sex (odd ratio [OR] = 1.974, 95% confidence interval [CI], 1.243-2.774; P =0.004), age < 45 years (OR = 4.621, 95% CI, 2.160-9.347; P < 0.001), US-reported CLN status (OR = 1.894, 95% CI, 0.754-3.347; P =0.005), multifocality (OR = 1.793, 95% CI, 0.774-2.649; P =0.007), tumor size ≥ 0.6cm (OR = 1.731, 95% CI,0.793-3.852; P =0.018), ETE (OR = 3.772, 95% CI, 1.752-8.441;P< 0.001) and microcalcification (OR = 2.316, 95% CI, 1.099-4.964; P < 0.001) were taken into account. The predictive nomogram was established by involving all the factors above used for preoperative prediction of CLNM in patients with PTCM. The nomogram model showed an AUC of 0.839 and an accuracy of 77.9% in predicting CLNM. Furthermore, the calibration curve demonstrated a strong consistency between nomogram and clinical findings in prediction CLNM for PTMC.

CONCLUSIONS

The nomogram achieved promising results for predicting preoperative CLNM in PTMC by combining clinical and US risk factor. Our proposed prediction model is able to help determine an individual's risk of CLNM in PTMC, thus facilitate reasonable therapy decision making.

Computed Tomography-Based Radiomics Model to Predict Central Cervical Lymph Node Metastases in Papillary Thyroid Carcinoma: A Multicenter Study

OBJECTIVES

This study aimed to develop a computed tomography (CT)-based radiomics model to predict central lymph node metastases (CLNM) preoperatively in patients with papillary thyroid carcinoma (PTC).

METHODS

In this retrospective study, 678 patients with PTC were enrolled from Yantai Yuhuangding Hot3spital (n=605) and the Affiliated Hospital of Binzhou Medical University (n=73) within August 2010 to December 2020. The patients were randomly divided into a training set (n=423), an internal test set (n=182), and an external test set (n=73). Radiomics features of each patient were extracted from preoperative plain scan and contrast-enhanced CT images (arterial and venous phases). One-way analysis of variance (ANOVA) and least absolute shrinkage and selection operator algorithm were used for feature selection. The K-nearest neighbor, logistics regression, decision tree, linear-support vector machine (linear-SVM), Gaussian-SVM, and polynomial-SVM algorithms were used to establish radiomics models for CLNM prediction. The clinical risk factors were selected by ANOVA and multivariate logistic regression. Incorporated with clinical risk factors, a combined radiomics model was established for the preoperative prediction of CLNM in patients with PTCs. The performance of the combined radiomics model was evaluated using the receiver operating characteristic (ROC) and calibration curves in the training and test sets. The clinical usefulness was evaluated through decision curve analysis (DCA).

RESULTS

A total of 4227 radiomic features were extracted from the CT images of each patient, and 14 non-zero coefficient features associated with CLNM were selected. Four clinical variables (sex, age, tumor diameter, and CT-reported lymph node status) were significantly associated with CLNM. Linear-SVM led to the best prediction model, which incorporated radiomic features and clinical risk factors. Areas under the ROC curves of 0.747 (95% confidence interval [CI] 0.706-0.782), 0.710 (95% CI 0.634-0.786), and 0.764 (95% CI 0.654-0.875) were obtained in the training, internal, and external test sets, respectively. The linear-SVM algorithm also showed better sensitivity (0.702 [95% CI 0.600-0.790] vs. 0.477 [95% CI 0.409-0.545]) and accuracy (0.670 [95% CI 0.600-0.738] vs. 0.642 [95% CI 0.569-0.712]) than an experienced radiologist in the internal test set in the combined radiomics model. The calibration plot reflected a favorable agreement between the actual and estimated probabilities of CLNM. The DCA indicated the clinical usefulness of the combined radiomics model.

CONCLUSION

The combined radiomics model is a non-invasive preoperative tool that incorporates radiomic features and clinical risk factors to predict CLNM in patients with PTC.