Nomogram to differentiate benign and malignant thyroid nodules in the American College of Radiology Thyroid Imaging Reporting and Data System level 5

Diagnostic Nomogram Model for ACR TI-RADS 4 Nodules Based on Clinical, Biochemical Data and Sonographic Patterns

OBJECTIVES

The objective of this study was to develop and validate a nomogram model integrating clinical, biochemical and ultrasound features to predict the malignancy rates of Thyroid Imaging Reporting and Data System 4 (TR4) thyroid nodules.

METHODS

A total of 1557 cases with confirmed pathological diagnoses via fine-needle aspiration (FNA) were retrospectively included. Univariate and multivariate logistic regression analyses were conducted to identify independent predictors of malignancy. These predictors were incorporated into the nomogram model, and its predictive performance was evaluated using receiver-operating characteristic curve (AUC), calibration plots, net reclassification improvement (NRI), integrated discrimination improvement (IDI) and decision curve analysis (DCA).

RESULTS

Eight out of 22 variables-age, margin, extrathyroidal extension, halo, calcification, suspicious lymph node metastasis, aspect ratio and thyroid peroxidase antibody-were identified as independent predictors of malignancy. The calibration curve demonstrated excellent performance, and DCA indicated favourable clinical utility. Additionally, our nomogram exhibited superior predictive ability compared to the current American College of Radiology (ACR) score model, as indicated by higher AUC, NRI, IDI, negative likelihood ratio (NLR) and positive likelihood ratio (PLR) values.

CONCLUSIONS

The developed nomogram model effectively predicts the malignancy rate of TR4 thyroid nodules, demonstrating promising clinical applicability.

Optimizing thyroid AUS nodules malignancy prediction: a comprehensive study of logistic regression and machine learning models

Background

The accurate diagnosis of thyroid nodules with indeterminate cytology, particularly in the atypia of undetermined significance (AUS) category, remains challenging. This study aims to predict the risk of malignancy in AUS nodules by comparing two machine learning (ML) and three conventional logistic regression (LR) models.

Methods

A retrospective study on 356 AUS nodules in 342 individuals from 6728 patients who underwent thyroid surgery in 2021. All the clinical, ultrasonographic, and molecular data were collected and randomly separated into training and validation cohorts at a ratio of 7: 3. ML (random forest and XGBoost) and LR (lasso regression, best subset selection, and backward stepwise regression) models were constructed and evaluated using area under the curve (AUC), calibration, and clinical utility metrics.

Results

Approximately 90% (321/356) of the AUS nodules were malignant, predominantly papillary thyroid carcinoma with 68.6% BRAF V600E mutations. The final LR prediction model based on backward stepwise regression exhibited superior discrimination with AUC values of 0.83 (95% CI: 0.73-0.92) and 0.80 (95% CI: 0.67-0.94) in training and validation, respectively. Well calibration, and clinical utility were also confirmed. The ML models showed moderate performance. A nomogram was developed on the final LR model.

Conclusions

The LR model developed using the backward stepwise regression, outperformed ML models in predicting malignancy in AUS thyroid nodules. The corresponding nomogram based on this model provides a valuable and practical tool for personalized risk assessment, potentially reducing unnecessary surgeries and enhancing clinical decision-making.

A nomogram based on clinicopathological and ultrasound characteristics to predict central neck lymph node metastases in papillary thyroid cancer

Purpose

Papillary thyroid cancer (PTC) has grown rapidly in prevalence over the past few decades, and central neck lymph node metastasis (CNLNM) is associated with poor prognoses. However, whether to carry out preventive central neck lymph node dissection (CNLND) is still controversial. We aimed to construct a prediction model of CNLNM to facilitate making clinical surgical regimens.

Methods

A total of 691 patients with PTC between November 2018 and December 2021 were included in our study. Univariate and multivariate analyses were performed on basic information and clinicopathological characteristics, as well as ultrasound characteristics (American College of Radiology (ACR) scores). The prediction model was constructed and performed using a nomogram, and then discriminability, calibrations, and clinical applicability were evaluated.

Results

Five variables, namely, male, age >55 years, clinical lymph node positivity, tumor size ≥1 cm, and ACR scores ≥6, were independent predictors of CNLNM in the multivariate analysis, which were eventually included to construct a nomogram model. The area under the curve (AUC) of the model was 0.717, demonstrating great discriminability. A calibration curve was developed to validate the calibration of the present model by bootstrap resampling, which indicated that the predicted and actual values were in good agreement and had no differentiation from the ideal model. The decision curve analysis (DCA) indicated that the prediction model has good clinical applicability.

Conclusions

Our non-invasive prediction model combines ACR scores with clinicopathological features presented through nomogram and has shown good performance and application prospects for the prediction of CNLNM in PTCs.