Related Experiment Video
Updated: Sep 28, 2025
03:55
Computer-Aided Three-Dimensional Visualization in the Treatment of Locally Advanced Thyroid Cancer
Published on: June 9, 2023
637
Preoperative comprehensive malignancy risk estimation for thyroid nodules: Development and verification of a
Guangdong Shao1, Baoqi Sun2, Mingming Shi1
1Department of Thyroid and Breast Diagnosis and Treatment Center, Weifang Hospital of Traditional Chinese Medicine, No. 1055 Weizhou Road, Kuiwen District, Weifang City, 261000, Shandong Province, China.
Summary
A new web-based model predicts thyroid nodule malignancy risk using imaging and lab data. This tool helps avoid unnecessary treatments by accurately stratifying nodules before surgery.
Area of Science:
- Endocrinology
- Oncology
- Medical Imaging
Background:
- Thyroid nodules require accurate diagnosis to prevent overtreatment.
- Distinguishing benign from malignant nodules pre-surgically is clinically significant.
Purpose of the Study:
- To develop and validate a comprehensive web-based predictive model for thyroid nodule malignancy risk.
- To integrate imaging and laboratory criteria for improved diagnostic accuracy.
Main Methods:
- Retrospective analysis of 682 patients with thyroid nodules.
- Development of a predictive model using binary logistic regression.
- Validation of the model on a separate dataset.
Main Results:
- Key predictors identified: TI-RADS, Bethesda categories, BRAF V600E mutation, Calcitonin, and FNA-Tg.
- A 10-grade risk scoring system demonstrated strong correlation with malignancy risk (2.06%-100%).
- High diagnostic performance with AUCs of 0.972 (development) and 0.946 (validation).
Conclusions:
- A simple, reliable web-based model effectively stratifies thyroid nodule malignancy risk.
- The model integrates diverse criteria for pre-operative risk assessment.
- This tool aids in clinical decision-making for thyroid nodule management.
