Related Experiment Video
Updated: Apr 20, 2026

A Swin Transformer-Based Model for Thyroid Nodule Detection in Ultrasound Images
Published on: April 21, 2023
Diagnosis of thyroid nodules using virtual touch tissue quantification value and anteroposterior/transverse diameter
Tao Li1, Pei Zhou2, Xuming Zhang1
1School of Life Science and Technology, Huazhong University of Science and Technology, Wuhan, HuBei Province, People's Republic of China.
Abstract:
This paper proposes an effective approach to differential diagnosis of thyroid nodules using a hierarchical classification model based on the Virtual Touch tissue quantification (VTQ) value and anteroposterior/transverse diameter (A/T) ratio. One hundred twenty nodules (92 benign, 28 malignant) were analyzed using this approach by combining the quantitative elastic characteristic with the conventional sonographic feature. First, nodules were classified as benign (VTQ values <2.27 m/s), malignant (VTQ values >2.73 m/s) and indeterminate (2.27 m/s ≤ VTQ values ≤2.73 m/s) using two cutoff points selected on the basis of receiver operating characteristic analysis. Second, the indeterminate nodules were separated into malignant and benign nodules using an A/T ratio ≥1. The advantage of this approach was that it could alleviate the limitation of an overlap in VTQ values between benign and malignant nodules. According to the pathologic results, the accuracy of this approach was 95%. The proposed approach may potentially improve diagnostic accuracy.

