Registered title, quoted from ClinicalTrials.gov
Construction of a Benchmark for Breast Ultrasound AI Interpretation and Performance Evaluation of Multimodal AI Models
An open trial on breast cancer, registered on ClinicalTrials.gov. Led by Peking Union Medical College Hospital; 0 other hospitals on this site are listed as study sites.
Status, phase and start date as registered on ClinicalTrials.gov on 18 Nov 2026; the registry may have changed since. Whether someone can take part is decided by the study team. Not medical advice.
The study team’s contact details are on the registry record. This site does not pass on enquiries.
Hospitals in mainland China listed for this trial 1
Hospitals on this site that the registry record names as lead sponsor or study site, where the site is open. Lead sponsor first, then study sites A–Z.
The record may name more sites, including outside mainland China, and sites can open or close during a trial. Full list of sites on ClinicalTrials.gov
Source: ClinicalTrials.gov record NCT07500428, read on 18 Nov 2026; hospital names matched to this site’s hospital list. How names are matched
Checklist for this trial 2
Only the points this trial’s record answers definitely: the registered age and sex, and criteria lines that state a number, score, stage or named test, quoted word for word. Tick a point once it has been compared with the medical records.
Show the checklist (2 points)Hide the checklist
18 years to 75 years; sex: women only
Registered eligibility fields
Pathological diagnosis confirmed (for benign and malignant lesion groups), or normal breast status confirmed by a senior radiologist with >15 years of breast ultrasound experience (for the normal group)
Inclusion criterion 3
Ticks stay on this page; they are not saved or sent. Meeting every point is not enough: the study team decides after its own screening. Lines that need the team’s judgement, such as “sufficient organ function”, are not in this list; they are in the full criteria below. Not stated definitely in this record: stage or severity, as the trial defines it; earlier treatment; day-to-day activity level; blood tests: counts, liver and kidney function; scans and measurable lesions; other conditions the trial excludes; pregnancy, breastfeeding and contraception.
Questions to ask the study team
Picked from the record by fixed keyword rules (not by AI, not checked by hand) and quoted from ClinicalTrials.gov, NCT07500428, read on 10 Nov 2026.
Registered criteria
Quoted as registered on ClinicalTrials.gov, in English, read on 10 Nov 2026. This site has not checked, shortened or translated them. Meeting the written criteria is not enough: the study team decides after its own screening.
Inclusion Criteria:
- B-mode breast ultrasound grayscale images from the institutional PACS database or from published open-access breast ultrasound datasets with documented original institutional ethics approval
- Image quality adequate for clinical diagnosis with clear visualization of the region of interest
- Pathological diagnosis confirmed (for benign and malignant lesion groups), or normal breast status confirmed by a senior radiologist with >15 years of breast ultrasound experience (for the normal group)
- Complete de-identification with removal of all personally identifiable information
Exclusion Criteria:
- Severely degraded image quality precluding meaningful BI-RADS assessment
- Duplicate images from the same patient (only the most representative image retained per lesion)
- Images with residual personally identifiable information after de-identification processing
- Cases with ambiguous, disputed, or unavailable pathological results
- Non-B-mode ultrasound images, including elastography, contrast-enhanced ultrasound, and Doppler imaging
Quoted from ClinicalTrials.gov, NCT07500428, participation criteria. Read on 10 Nov 2026.
Where this information comes from
The registry record
ClinicalTrials.gov (US National Library of Medicine), record NCT07500428, read on 18 Nov 2026. A trial has a page here while its overall status is recruiting, not yet recruiting or enrolling by invitation and at least one listed hospital’s site is open. Trials registered only in China’s own registry (ChiCTR) are not included.
Diseases and hospitals
The trial is listed under a disease when this site’s ClinicalTrials.gov search for that disease in China returns it. Sponsor and site names in the record are matched to this site’s hospital list (ROR identifiers, checked aliases). Definitions
What is left out
Names of investigators and contact persons in the record, and anything not in the record. The title and criteria are quoted from the record.