Construction of a Benchmark for Breast Ultrasound AI Interpretation and Performance Evaluation of Multimodal AI Models

Conditions studied

Breast Neoplasms, Breast Diseases, Ultrasonography

About this study

This single-center, retrospective, observational study aims to construct a standardized benchmark evaluation system for intelligent breast ultrasound image interpretation and to systematically assess the diagnostic performance of current mainstream multimodal artificial intelligence (AI) models. De-identified B-mode breast ultrasound images with confirmed pathological diagnoses will be retrospectively collected from the institutional archive (2018-2025) and supplemented with images from published open-access datasets. Expert radiologists with varying experience levels will independently annotate all images according to the American College of Radiology (ACR) Breast Imaging Reporting and Data System (BI-RADS) v2025 criteria, including glandular tissue composition, lesion characterization (mass vs. non-mass lesion), morphological descriptors, and final BI-RADS classification. Baseline deep learning models (CNN-based ResNet-50 and Transformer-based USFM) will be trained to establish performance baselines and to stratify cases by diagnostic difficulty through cross-architecture consensus. Multiple multimodal large language models (MLLMs), including both general-purpose and medical-domain models, will then be evaluated via standardized API calls using BI-RADS-guided chain-of-thought prompts at temperature 0 for reproducibility. Primary endpoints include BI-RADS classification accuracy and diagnostic AUC for benign-malignant differentiation. Model robustness and safety will be assessed through out-of-distribution rejection testing, temperature-stability experiments, and thinking-mode ablation studies. This study adheres to the FLAIR and TRIPOD-LLM reporting guidelines.

Interventions

• Diagnostic Test: Multimodal AI Model Diagnostic Evaluation — Retrospective evaluation of de-identified breast ultrasound images by multiple AI systems, including baseline deep learning models (ResNet-50, USFM) and multimodal large language models, using standardized BI-RADS-guided chain-of-thought prompts via API. No patient contact or clinical decision-making is involved.

Primary outcomes

• Diagnostic Accuracy for Pathological Diagnosis (At study completion, approximately 12 months)
• BI-RADS Classification Accuracy (At study completion, approximately 12 months)

Eligibility information

Study locations

• Peking Union Medical College Hospital, Beijing, 100730 China