Related Experiment Video
Updated: Apr 1, 2026
04:23
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
Published on: April 21, 2023
2.4K
Web-Based Malignancy Risk Estimation for Thyroid Nodules Using Ultrasonography Characteristics: Development and
Young Jun Choi1, Jung Hwan Baek1, Seung Hee Baek2
11 Department of Radiology and Research Institute of Radiology, University of Ulsan College of Medicine , Seoul, Korea.
Thyroid : Official Journal of the American Thyroid Association
|October 7, 2015
Summary
A new web-based model simplifies thyroid nodule analysis using ultrasound features to predict malignancy risk. This tool aids clinicians in stratifying nodules, improving diagnostic accuracy and patient care.
Area of Science:
- Radiology
- Oncology
- Medical Informatics
Background:
- Clinical analysis of thyroid nodules requires simplified, practical prediction models.
- Ultrasonography is a key imaging modality for thyroid nodule assessment.
Purpose of the Study:
- To construct and validate a user-friendly, web-based predictive model for thyroid nodule malignancy risk stratification.
- To utilize ultrasonography characteristics for improved clinical decision-making.
Main Methods:
- Radiologists assessed ultrasonography images of thyroid nodules based on specific criteria (content, echogenicity, shape, margin, calcifications).
- Multivariate logistic regression was employed to identify predictors of malignancy.
- A development dataset (849 nodules) and a validation dataset (453 nodules) were used.
Main Results:
- Key ultrasonography features predicting malignancy included solid content, taller-than-wide shape, spiculated/ill-defined margins, and specific calcifications.
- A 14-point risk scoring system was developed, correlating with malignancy risk (3.8%–97.4%).
- The model demonstrated high predictive accuracy with areas under the ROC curve of 0.903 (development) and 0.897 (validation).
Conclusions:
- A simple, reliable web-based model effectively stratifies thyroid nodules by malignancy probability using ultrasonography features.
- The developed risk scoring system offers a practical tool for clinical application.
- This model enhances the ability to assess thyroid nodule risk in a clinical setting.
