Benchmark and performance test for AI reading breast ultrasound images

Is this trial a fit for you?

A plain-English reading of the study's rules. Only the study team can confirm whether you qualify.

What this trial studies

This retrospective single-center project will collect de-identified B-mode breast ultrasound images from Peking Union Medical College Hospital (2018–2025) and from open-access datasets to create a standardized evaluation benchmark. Expert radiologists with junior and senior experience will independently annotate images using ACR BI-RADS v2025 criteria, with a senior arbitrator resolving disagreements. Baseline CNN (ResNet-50) and transformer (USFM) models will be trained, and multiple multimodal large language models will be tested on an evaluation set plus an out-of-distribution safety test set. Images will be stratified into difficulty tiers based on cross-architecture model consensus to characterize where models succeed or fail.

Why it matters

Who can join: the full rules

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

Where it's running

• Peking Union Medical College Hospital
  Beijing, China Recruiting

Study contacts

• Principal investigator: Qingli Zhu, MD — Peking Union Medical College Hospital
• Study coordinator: Qingli Zhu, MD
• Email: zqlpumch@126.com
• Phone: +86 13621376699

Similar studies

Other studies for the same condition or treatment. Recruiting studies are listed first.

• Dalpiciclib plus hormone therapy for early HR+/HER2- breast cancer

  This trial tests whether adding dalpiciclib at two different doses and durations to standard hormone therapy helps people with early HR+/HER2- breast cancer.

  Recruiting Phase 2 Fujian Cancer Hospital · Fuzhou, Fujian and 1 other locations · NCT07581834

• How local breast cancer treatments affect side effects and quality of life in Chinese patients

  This project will test how surgery and related treatments (chemotherapy, radiation, hormone or targeted therapy) affect side effects and daily quality of life for people with early-stage breast cancer in China.

  Recruiting Cancer Institute and Hospital, Chinese Academy of Medical Sciences · Beijing, Beijing Municipality and 2 other locations · NCT07010939

• Predicting outcomes for advanced breast cancer patients treated with CDK4/6 inhibitors and endocrine therapy

  This study is trying to see how well different CDK4/6 inhibitors combined with hormone therapy work for people with advanced breast cancer and how safe they are.

  Recruiting at 24 sites Fondazione IRCCS Istituto Nazionale dei Tumori, Milano · Aviano and 23 other locations · NCT06805812

• Understanding breast cancer risk factors and outcomes in Florida

  This study is trying to understand the different risk factors and outcomes of breast cancer in men and women in Florida by collecting samples and survey information from new patients at breast clinics.

  Recruiting at 2 sites Memorial Sloan Kettering Cancer Center · Los Angeles, California and 2 other locations · NCT06780176

• 68Ga-FAPI-46 versus 18F-FDG PET imaging in ER-positive breast cancer

  This test compares whether 68Ga-FAPI-46 PET scans find more cancer spread than standard 18F-FDG PET scans in people with ER-positive breast cancer who need staging.

  Recruiting Phase 2 Maastricht University Medical Center · Maastricht · NCT06335069

• Combination treatment for brain metastasis from ER+/HER-2- breast cancer

  This study is testing a new combination treatment of elacestrant and abemaciclib to see if it can help people with brain metastases from estrogen receptor-positive, HER-2 negative breast cancer.

  Recruiting at 78 sites Phase 1/2 Stemline Therapeutics, Inc. · Fullerton, California and 85 other locations · NCT05386108

How to take part

1. Review the eligibility criteria above with your treating physician.
2. Check the official listing on ClinicalTrials.gov for the latest contact details and recruitment status.
3. Contact the study coordinator or principal investigator to ask for pre-screening. Pre-screening is free and never obligates you to enroll.