Related Experiment Video
Updated: Oct 14, 2025

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
Published on: April 21, 2023
Predictive Value of a Genomic Classifier in Indeterminate Thyroid Nodules Based on Nodule Size
Jared C Dublin1, Michael Papazian1, Elcin Zan2
1New York University Grossman School of Medicine, New York.
Genomic classifier tests for indeterminate thyroid nodules (ITNs) show improved accuracy with increasing nodule size. Larger ITNs with negative results have a low cancer risk, potentially supporting less invasive initial management.
Area of Science:
- Endocrinology
- Oncology
- Genetics
Background:
- Genomic classifiers aid in managing indeterminate thyroid nodules (ITNs).
- The impact of ITN size on diagnostic accuracy of genomic classifiers remains understudied.
Purpose of the Study:
- To evaluate the diagnostic performance of a genomic classifier (ThyroSeq) in relation to ITN size.
- To assess the association between genetic signatures and cancer risk in ITNs.
Main Methods:
- A retrospective case series of 218 ITNs from patients undergoing genomic classifier testing and surgery.
- Analysis of diagnostic metrics (PPV, NPV, sensitivity, specificity) stratified by ITN size (<2 cm, 2-4 cm, >4 cm).
Main Results:
- The overall cancer prevalence was 23.9%.
- Positive predictive value (PPV) and negative predictive value (NPV) improved with increasing ITN size (e.g., PPV 25% and NPV 79% for <2 cm vs. PPV 50% and NPV 89% for >4 cm).
- Test specificity was significantly higher for larger ITNs (40% for >4 cm vs. 15% for <2 cm, P=.01).
Conclusions:
- ThyroSeq test performance, particularly specificity, improves with larger indeterminate thyroid nodule size.
- Large ITNs with negative genomic classifier results have a low risk of malignancy.
- Initial thyroid lobectomy may be a sufficient management strategy for genomic classifier-negative ITNs >4 cm.
More Related Videos
07:13Comparison of Predictive Performance of Three Lymph Node Staging Systems in Colorectal Signet Ring Cell Carcinoma Based on Machine Learning Model
Published on: April 18, 2025
10:26Author Spotlight: A 3D Digital Model for the Diagnosis and Treatment of Pulmonary Nodules
Published on: May 19, 2023