Related Experiment Video
Updated: Jan 3, 2026
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
Published on: April 21, 2023
Thyroid Image Reporting and Data System Categorization: Effectiveness in Pediatric Thyroid Nodule Assessment
Cigdem Uner1, Sonay Aydin, Berna Ucan
1Department of Radiology, Dr Sami Ulus Training and Research Hospital, Ankara, Turkey.
Insights
The Thyroid Image Reporting and Data System (TI-RADS) effectively assesses pediatric thyroid nodules. Categories 4 and 5 indicate the highest malignancy risk, warranting pathological evaluation.
Area of Science:
- Pediatric Endocrinology
- Radiology
- Oncology
Background:
- Thyroid nodules are less common in children but carry a significant malignancy risk (approx. 25%).
- Thyroid Image Reporting and Data System (TI-RADS) validation in pediatric cases is limited, despite its use in adults.
Purpose of the Study:
- To evaluate the diagnostic performance of the TI-RADS risk stratification method for pediatric thyroid nodules.
- To determine the efficacy of TI-RADS in predicting malignancy in pediatric thyroid nodules.
Main Methods:
- Retrospective analysis of 68 nodules from 64 pediatric patients (aged ≥18 years) with pathological diagnoses.
- Evaluation of ultrasonography (US) images or reports to assign TI-RADS scores and categories.
- Statistical analysis including Area Under the Curve (AUC) estimation.
Main Results:
- The study included 68 nodules (48 benign, 20 malignant) from 64 patients (mean age 15.15 years).
- TI-RADS categorization demonstrated significant diagnostic efficacy with an AUC of 0.89 (95% CI: 0.80-0.98).
- TI-RADS categories 4 and 5 were identified as optimal cutoffs for predicting malignancy.
Conclusions:
- TI-RADS is an effective tool for assessing pediatric thyroid nodules.
- Pediatric thyroid nodules classified as TI-RADS 4 or 5 exhibit the highest risk of malignancy and require pathological examination.
Abstract:
Ultrasonography (US) is the main imaging method to assess thyroid nodules. In the pediatric population, thyroid nodules are less prevalently seen than those in adults. However, approximately 25% of the nodules can be malignant. Validation of Thyroid Image Reporting and Data System (TI-RADS) classification has been mostly tested in adults, and information on pediatric cases is limited. In the current study, we aim to define the diagnostic power of the TI-RADS risk stratification method in pediatric thyroid nodules.The study population consists of 68 nodules of 64 patients (20 malignant, 48 benign). We have included patients at least 18 years and with a pathological diagnosis. The researchers evaluated the US images of the patients, if available. If US images cannot be found, US reports of the patients were evaluated to define TI-RADS points and category.Mean age of the population is 15.15 ± 2.66 years. There were 10 (17.7%) male patients and 58 (85.3%) female patients. Median nodule size was 8 mm. Final diagnosis was benign in 48 nodules (70.6%) and malignant in 20 nodules (29.4%). There were 5 follicular carcinoma cases and 15 papillary carcinoma cases. The area under the curve estimate was 0.89 (95% confidence interval, 0.80-0.98), and the area under the curve value confirmed the diagnostic efficacy of TI-RADS categorization in pediatric thyroid nodules. A TI-RADS category of 4 or 5 seemed to be a good cutoff point to predict malignancy.To conclude, TI-RADS categorization can be effectively used to assess pediatric thyroid nodules. Nodules with TI-RADS categories of 4 and 5 have the greatest risk of malignancy, and they should be evaluated pathologically.
