Artificial intelligence-enabled ultrasound diagnosis and stratification of follicular thyroid neoplasms: a multi-center study
Preoperatively distinguishing follicular thyroid carcinoma (FTC) from follicular thyroid adenoma (FTA) remains a significant clinical challenge. Current ultrasound risk stratification systems show limited efficacy for follicular neoplasms, and existing artificial intelligence (AI) approaches lack sufficient validation. We developed and validated a deep learning model using ultrasound images to differentiate FTC from FTA and classify FTC into invasion subtypes. This multicenter retrospective study incorporated data from 31 hospitals, using 1531 patients for model development and 900 across three external test sets for validation. The model demonstrated high diagnostic performance, with AUCs of 0.816-0.847 for FTC vs FTA discrimination across external test sets and robust performance across subtypes (AUC range 0.754-0.910), and generalized well to varied clinical settings. Triple-classification macro-AUCs were 0.818-0.861. It consistently outperformed radiologists and improved diagnostic accuracy as an assistive tool. Our AI model provides a reliable, non-invasive tool for preoperative diagnosis and risk stratification of follicular thyroid neoplasms.
Authors
- Li Pz
- Weide Dai
- Linyi Ma (ORCID: https://orcid.org/0009-0007-3945-6303)
- Runqin Miao
- MD Ligang Cui
- Man Lu (ORCID: https://orcid.org/0000-0001-6339-229X)
- Xiang Xie (ORCID: https://orcid.org/0009-0002-6535-3045)
- Shengnan Huo
- Xinping Wu (ORCID: https://orcid.org/0000-0002-6330-1615)
- Haoyan Zhang (ORCID: https://orcid.org/0000-0003-3067-0710)
- Jianming Li (ORCID: https://orcid.org/0000-0002-2585-4753)
- Ying Liu
- Liping Yang (ORCID: https://orcid.org/0000-0002-2896-1474)
- Guoyong Hua
- Changsong Xu (ORCID: https://orcid.org/0000-0003-3843-5096)
- B. Ping
- Kexin Lou
- Yuancheng Cang
- Fang Mei
- Jie Yu
- Huan Zheng
- Baoding Chen
- Kun Wang
- Ping Liang
- Lin xue Qian
- Xin Yang
- Yong Xu
- Liping Liu
Institutions
- Handan College (CN)
- Jiangsu University (CN)
- Nanjing University of Chinese Medicine (CN)
- University of Electronic Science and Technology of China (CN)
- Shanxi Medical University (CN)
- Xuzhou Medical College (CN)
- Capital Medical University (CN)
- Anhui Medical University (CN)
- Chinese Academy of Sciences (CN)
- Peking University (CN)
- Tianjin Medical University Cancer Institute and Hospital (CN)
- Chinese PLA General Hospital (CN)
- Second Affiliated Hospital of Guangzhou Medical University (CN)
- Fudan University Shanghai Cancer Center (CN)
- Shandong Institute of Automation (CN)
- Huaian First People’s Hospital (CN)
- Beijing Academy of Artificial Intelligence (CN)
- Shanxi Provincial Cancer Hospital (CN)
- Affiliated Hospital of Jiangsu University (CN)
- Sichuan Cancer Hospital (CN)
- Beijing Hospital (CN)
- Second People’s Hospital of Huai’an (CN)
- First Hospital of Shanxi Medical University (CN)
- Peking University Shenzhen Hospital (CN)
- Xuzhou Central Hospital (CN)
- Peking University Third Hospital (CN)
- Beijing Friendship Hospital (CN)
- Second Affiliated Hospital of Anhui Medical University (CN)
- Qinghai Provincial Peoples Hospital (CN)
- Guangdong Provincial People's Hospital (CN)
- University of Chinese Academy of Sciences (CN)
- Anyang Hospital of Traditional Chinese Medicine (CN)
- Southern Medical University (CN)
- Nanjing Medical University (CN)
- China Medical University (CN)
- Guangzhou Medical University (CN)
Publication Details
- Journal
- npj Digital Medicine
- Published
- 2026-03-05
- DOI
- https://doi.org/10.1038/s41746-026-02489-6
- Primary Topic
- Thyroid Cancer Diagnosis and Treatment
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Natural Science Foundation of China