Download PDF
Exp Clin Endocrinol Diabetes 2023; 131(10): 508-514
DOI: 10.1055/a-2122-5585
Article

Stratifying High-Risk Thyroid Nodules Using a Novel Deep Learning System

Chia-Po Fu
1   Graduate Institute of Biomedical Electronics and Bioinformatics, College of Electrical Engineering and Computer Science, National Taiwan University, Taipei, Taiwan
2   Division of Endocrinology and Metabolism, Department of Medicine, Taichung Veterans General Hospital, Taichung, Taiwan
3   Department of Medicine, Chung Shan Medical University, Taichung, Taiwan
4   Department of Medicine, National Yang Ming Chiao Tung University, Taipei, Taiwan
,
Ming-Jen Yu
1   Graduate Institute of Biomedical Electronics and Bioinformatics, College of Electrical Engineering and Computer Science, National Taiwan University, Taipei, Taiwan
,
Yao-Sian Huang
5   Department of Computer Science and Information Engineering, National Changhua University of Education, Changhua County, Taiwan
,
Chiou-Shann Fuh
1   Graduate Institute of Biomedical Electronics and Bioinformatics, College of Electrical Engineering and Computer Science, National Taiwan University, Taipei, Taiwan
6   Department of Computer Science and Information Engineering, National Taiwan University, Taipei, Taiwan
,
Ruey-Feng Chang
1   Graduate Institute of Biomedical Electronics and Bioinformatics, College of Electrical Engineering and Computer Science, National Taiwan University, Taipei, Taiwan
6   Department of Computer Science and Information Engineering, National Taiwan University, Taipei, Taiwan
› Author Affiliations Funding This work was supported by grants from the Ministry of Science and Technology of Taiwan (MOST 110–2221-E-002–122-MY3 and MOST 110–2221-E-002–123-MY3).
› Further Information
Also available at
    Permissions and Reprints

Abstract

Introduction The current ultrasound scan classification system for thyroid nodules is time-consuming, labor-intensive, and subjective. Artificial intelligence (AI) has been shown to increase the accuracy of predicting the malignancy rate of thyroid nodules. This study aims to demonstrate the state-of-the-art Swin Transformer to classify thyroid nodules.

Materials and Methods Ultrasound images were collected prospectively from patients who received fine needle aspiration biopsy for thyroid nodules from January 2016 to June 2021. One hundred thirty-nine patients with malignant thyroid nodules were enrolled, while 235 patients with benign nodules served as controls. Images were fed to Swin-T and ResNeSt50 models to classify the thyroid nodules.

Results Patients with malignant nodules were younger and more likely male compared to those with benign nodules. The average sensitivity and specificity of Swin-T were 82.46% and 84.29%, respectively. The average sensitivity and specificity of ResNeSt50 were 72.51% and 77.14%, respectively. Receiver operating characteristics analysis revealed that the area under the curve of Swin-T was higher (AUC=0.91) than that of ResNeSt50 (AUC=0.82). The McNemar test evaluating the performance of these models showed that Swin-T had significantly better performance than ResNeSt50.

Swin-T classifier can be a useful tool in helping shared decision-making between physicians and patients with thyroid nodules, particularly in those with high-risk characteristics of sonographic patterns.

Key words

thyroid nodule, sonography, deep learning, ResNeSt50, Swin-T


Publication History

Received: 26 March 2023
Received: 15 June 2023

Accepted: 28 June 2023

Article published online:
21 August 2023

© 2023. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany