+1 877 705 1914• 24 / 7
HIPAA Compliant
ISO 27001 Certified

Actively Recruiting

Age: 18Years - 80Years
All Genders
ID07488325

Using Ultrasound and Cytology Images to Improve Malignancy Detection in Bethesda III Thyroid Nodules

Led by Sun Yat-Sen Memorial Hospital of Sun Yat-Sen University · Updated on 2026-04-01

396

Participants Needed

1

Research Sites

2 weeks

Total Duration

AI-Summary

What This Trial Is About

Researchers are working to develop a multimodal deep learning model that combines ultrasound images and cytology whole-slide images to better predict the risk of malignancy in Bethesda III thyroid nodules. This study focuses on patients who have had fine-needle aspiration biopsies of thyroid nodules categorized as Bethesda III, with the goal of improving diagnostic accuracy for these uncertain cases. The study involves building and testing three types of models one based on ultrasound images, one on cytology images, and a fusion model that integrates both image types. The patients included in this research underwent conventional ultrasound exams and fine-needle aspiration between 2016 and 2024. The final diagnosis is confirmed either by postoperative histopathology or testing for the BRAFV600E mutation. Participants ultrasound and cytology images will be analyzed, and their surgical pathology results will be used as the reference standard. The main outcome measured is the pathological diagnosis confirmed one month after surgery. The researchers will assess how well the models predict malignancy using statistical methods and evaluate their usefulness in clinical decision-making. The study is observational and does not involve experimental treatments.

CONDITIONS

Brief Title

Multimodal Assessment of Malignancy in Atypia of Undetermined Significance in Thyroid Nodules Using Ultrasound and Cytology Whole-Slide Images

Research Team

J

Jingliang Ruan, Ph.D.

X

Xinmin Xiao, B.S.

Not the Right Trial for You?

Explore thousands of other clinical trials that might be a better match.
Sign up to get personalized trial recommendations delivered to your inbox.

Already have an account? Log in here