Skip to content

A Study of Artificial Intelligence-Assisted Diagnosis and Management of Thyroid Nodules

Clinical Utility of a Full-Process AI Diagnostic System for Clinical Decision Support in the Management of Thyroid Nodules: A Multicenter, Prospective, Randomized, Parallel-Controlled Trial

Status
Active, not recruiting
Phases
Unknown
Study type
Interventional
Source
ChiCTR
Registry ID
ChiCTR2600127481
Enrollment
Unknown
Registered
2026-07-01
Start date
2026-07-01
Completion date
Unknown
Last updated
2026-07-13

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

thyroid nodule?thyroid cancer

Interventions

control group:Routine checkup
experimental group:AI-assisted diagnosis

Sponsors

Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to 75 Years

Inclusion criteria

Inclusion criteria: 1. Male or female patients aged 18-75 years; 2. Patients with a solid or predominantly solid-cystic thyroid nodule identified on baseline outpatient or inpatient ultrasound, with a maximum diameter >= 10 mm and a C-TIRADS category of 4A or above; 3. If multiple eligible nodules are present, the most suspicious nodule will be selected; if nodules have the same risk category, the largest nodule will be chosen; 4. Patients who, based on clinical assessment, have indications for FNAB risk stratification and/or preoperative surgical evaluation; 5. No previous history of thyroid cancer, and no prior FNA, thyroid surgery, ablation, radioactive iodine (I-131) therapy, or head and neck radiotherapy; 6. Ultrasound images of adequate quality for AI system analysis, with pathological or follow-up reference standard verification expected to be completed within a short period. Adequate image quality is defined as clear visualization of the target nodule on both transverse and longitudinal views, with assessable key features including nodule size, location, margin, shape, internal echogenicity, calcification, vascularity, and relationship to surrounding tissues, and with the necessary measurement scale, orientation markers, and lesion localization information. When necessary, images of central and lateral cervical lymph nodes should also be acquired to meet the requirements of human interpretation, TI-RADS risk stratification, and AI system analysis; 7. Provision of written informed consent and ability to comply with all required clinical examinations and endpoint assessment within 6 months.

Exclusion criteria

Exclusion criteria: 1.Healthy individuals undergoing screening who do not have thyroid nodules (mandatory exclusion); 2.Any prior thyroid surgery, interventional procedure, or radiotherapeutic treatment; 3.Patients with a previously confirmed pathological diagnosis based on prior biopsy/FNA, or a history of thyroid malignancy; 4.Pregnancy or lactation, terminal malignancy, severe organ failure, or mental disorders that preclude tolerance of the required examinations; 5.Imaging data that do not meet study requirements, or refusal to comply with study procedures. Unqualified imaging includes, but is not limited to, unclear visualization of the target nodule, absence of key views such as the transverse or longitudinal plane, prominent image artifacts, incomplete measurement information, inability to perform TI-RADS risk stratification, or failure to meet the input and analysis requirements of the AI system. For cases with initially inadequate images, repeat image acquisition may be performed according to the standardized procedure. Subjects whose images still fail to meet the image quality control standards of this study after repeat acquisition will not be enrolled.

Design outcomes

Primary

MeasureTime frame
Overall decision accuracy;

Secondary

MeasureTime frame
Accuracy of surgery/biopsy/follow-up decisions;

Countries

China

Contacts

Public ContactWei Cai

Xinhua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine

caiwei@shsmu.edu.cn+86 21 6437 0045

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Jul 23, 2026