Related Experiment Video
Updated: Jan 2, 2026
04:23
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
Published on: April 21, 2023
2.2K
A Predictive Model to Distinguish Papillary Thyroid Carcinomas from Benign Thyroid Nodules Using Ultrasonographic
Da Fang1, Wenting Ma2, Lu Xu3
1Department of Endocrinology, The First Affiliated Hospital of Nanjing Medical University, Nanjing, Jiangsu, China (mainland).
Summary
A new ultrasound model accurately distinguishes papillary thyroid cancer (PTC) from benign nodules using nodule and lymph node features. This simple, cost-effective tool aids in early PTC diagnosis.
Area of Science:
- Radiology
- Oncology
- Medical Diagnostics
Background:
- Distinguishing papillary thyroid carcinoma (PTC) from benign thyroid nodules is crucial for effective patient management.
- Ultrasound (US) features of thyroid nodules and cervical lymph nodes are key indicators in this differentiation.
Purpose of the Study:
- To develop and validate a predictive model using US features to differentiate PTC from benign thyroid nodules.
- To assess the diagnostic performance of the developed model.
Main Methods:
- Retrospective analysis of preoperative US characteristics and postoperative histological data from 1119 thyroidectomy patients.
- Logistic regression modeling incorporating patient age, sex, thyroid nodule US features, and lymph node US features.
- Evaluation of the model's diagnostic accuracy using the area under the curve (AUC).
Main Results:
- Several US features, including hypoechogenicity, irregular shape, spiculate margin, taller-than-wide shape, microcalcifications, and capsular invasion, were independently associated with increased PTC risk.
- Lymph node metastasis features, such as absence of echogenic hilum and increased vascularization, also correlated with PTC.
- The developed risk score diagnosis system achieved an AUC of 0.916, indicating high diagnostic performance.
Conclusions:
- The developed predictive model is a reliable, simple, and cost-effective tool for diagnosing PTC.
- The model effectively utilizes US features of thyroid nodules and lymph nodes for accurate PTC prediction.
- This approach can aid clinicians in differentiating malignant from benign thyroid lesions.
