Related Experiment Video
Updated: Jul 26, 2025

04:23
A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
Published on: April 21, 2023
1.9K
Molecular testing raises thyroid nodule fine needle aspiration diagnostic value.
Dongyan Han1, Ding Min2, Rongli Xie3
1D Han, Department of pathology, Tongji University, Shanghai, China.
Endocrine Connections
|June 13, 2023
Summary
Molecular testing improves thyroid nodule diagnosis when fine needle aspiration biopsy (FNAB) is indeterminate. This 18-gene test identified key mutations, aiding in predicting malignancy and guiding treatment strategies.
Area of Science:
- Endocrinology
- Oncology
- Molecular Diagnostics
Background:
- Thyroid fine needle aspiration biopsy (FNAB) has limitations, with 16%-24% of cases yielding indeterminate results.
- Molecular testing offers a potential solution to enhance the diagnostic accuracy of FNAB for thyroid nodules.
Purpose of the Study:
- To investigate the gene mutation profile in patients with thyroid nodules.
- To assess the diagnostic performance of a novel 18-gene molecular test for thyroid nodules.
Main Methods:
- Analysis of 513 thyroid samples (414 FNABs, 99 FFPE) using a self-developed 18-gene molecular test.
- Calculation of diagnostic metrics including sensitivity, specificity, positive predictive value, negative predictive value, and accuracy.
Main Results:
- Detected 457 mutations in 428 samples, with BRAF mutations being the most frequent (73.3%).
- Evaluated diagnostic ability in Bethesda II and V-VI samples, showing improved accuracy with molecular testing compared to cytology alone.
- In Bethesda III-IV nodules, pathogenic mutations alone yielded a sensitivity of 76.2% and specificity of 66.7%.
Conclusions:
- Molecular testing, particularly an 18-gene panel, can significantly improve the diagnostic accuracy of indeterminate thyroid nodules.
- Understanding the molecular mechanisms at the genetic level is crucial for accurate malignancy prediction and developing tailored management strategies.

