ORIGINAL RESEARCH article
Front. Oncol.
Sec. Breast Cancer
Volume 15 - 2025 | doi: 10.3389/fonc.2025.1536365
This article is part of the Research TopicThe Essential Role of Multidisciplinary Teams in Breast Cancer Surgery: Collaboration for Superior Patient OutcomesView all 5 articles
Optimizing Breast Cancer Ultrasound Diagnosis: A Comparative Study of AI Model Performance and Image Resolution
Provisionally accepted
- 1The Second Clinical Medical College, Jinan University, Shen zhen, China
- 2Department of Interventional Radiology, shenzhen people'hospital, Shenzhen, China
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Objectives: To determine the optimal combination of artificial intelligence (AI) models and ultrasound (US) image resolutions for breast cancer diagnosis and evaluate whether this combination surpasses the diagnostic accuracy of senior radiologists. Materials and Methods: We systematically compared lightweight (MobileNet, Xception) and dense neural networks (ResNet50, DenseNet121) using three image resolutions (224×224, 320×320, 448×448 pixels). A retrospective cohort of 4,998 patients was divided into training/validation (8:2 ratio, n=3,578) and independent testing sets (n=1,410). Diagnostic performance was assessed via AUC, sensitivity, specificity, and analysis speed, with direct comparisons against senior radiologists. Results: MobileNet with 224×224 input achieved the highest AUC (0.924, 95%CI: 0.910-0.938) and accuracy (87.3%), outperforming senior US (AUC: 0.820, accuracy: 79.1%) and mammography doctors (AUC: 0.819, accuracy: 83.6%) (p<0.05). After excluding BI-RADS 4c and 5 nodules, diagnostic efficacy of MobileNet_224 is better than that of senior doctors (p<0.05), can reduce 60.1% false positive of US, and 46.6% of mammography. MobileNet_224 and MobileNet_320 had the fastest analyze speed. Conclusion: MobileNet_224 represents a novel, efficient AI framework for breast cancer diagnosis, demonstrating superior accuracy and speed compared to both complex AI models and experienced clinicians. This work highlights the critical role of optimizing model architecture and resolution to enhance diagnostic workflows and reduce unnecessary biopsies.
Keywords: artificial intelligence, breast cancer, diagnosis, Mammography, ultrasound
Received: 28 Nov 2024; Accepted: 07 May 2025.
Copyright: © 2025 Yin, Fang, Zhang and Shen. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence:
Yunqing Yin, The Second Clinical Medical College, Jinan University, Shen zhen, China
Wei Zhang, Department of Interventional Radiology, shenzhen people'hospital, Shenzhen, China
Xinying Shen, Department of Interventional Radiology, shenzhen people'hospital, Shenzhen, China
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.