乳腺病灶AI辅助临床决策软件的中美监管路径与临床研究探讨
收稿日期: 2025-11-04
修回日期: 2026-01-04
录用日期: 2026-05-26
网络出版日期: 2026-05-26
Comparative Exploration of Regulatory Pathways and Clinical Study for AI-Assisted Clinical Decision Support Software for Breast Lesion Diagnosis in China and the United States
Received date: 2025-11-04
Revised date: 2026-01-04
Accepted date: 2026-05-26
Online published: 2026-05-26
随着全球乳腺癌发病率的持续攀升及基层医疗资源的分布不均,人工智能(AI)技术在多模态乳腺影像分析中的临床应用价值日益凸显。本研究系统梳理了乳腺病灶的BI-RADS分级标准及钙化、肿块等核心临床特征,并深入剖析了传统X线摄影、超声及MRI等影像学手段在乳腺诊断中的局限性,阐述了AI与影像融合在图像增强、多模态诊疗及工作流自动化方面的临床赋能作用,重点聚焦美国食品药品管理局(FDA)对乳腺AI辅助临床决策类软件的分类监管要求,详细解构了不同申报路径下的临床证据体系,重点分析了辅助检测(CADe)、辅助诊断(CADx)及辅助分诊类产品的临床试验设计逻辑与评价指标差异。通过对比中美监管模式,结合我国最新的“人工智能+”行动意见,旨在为我国乳腺AI辅助临床决策类产品的审评要求提供理论支撑与实践范式,推动AI从单纯的高效工具进阶为医生的“智能临床伙伴”,最终实现优质医疗资源的公平可及。
郭佳莹
,
卢红霞
.
乳腺病灶AI辅助临床决策软件的中美监管路径与临床研究探讨
As the global incidence of breast cancer continues to rise and primary healthcare resources remain unevenly distributed, the clinical application value of artificial intelligence (AI) technology in multimodal breast imaging analysis has become increasingly prominent. This study systematically reviews the BI-RADS classification standards for breast lesions and core clinical characteristics, such as calcifications and masses, while providing an in-depth analysis of the limitations of traditional imaging modalities—including X-ray mammography, ultrasound, and MRI—in breast diagnosis. Furthermore, it elaborates on the clinical empowerment roles of AI-imaging fusion in image enhancement, multimodal diagnosis and treatment, and workflow automation. The study focuses on the classified regulatory requirements of the U.S. Food and Drug Administration (FDA) for breast AI-assisted clinical decision support software, deeply deconstructing the clinical evidence systems within various approval pathways. Specifically, it analyzes the differences in clinical trial design logic and evaluation metrics for computer-aided detection (CADe), computer-aided diagnosis (CADx), and computer-aided triage products. By comparing regulatory models in China and the United States, incorporating China's latest "AI+" action initiative, this study aims to provide theoretical support and practical paradigms for the evaluation requirements of breast AI-assisted clinical decision support products in China. The ultimate goal is to drive the advancement of AI from a mere efficient tool to an "intelligent clinical partner" for physicians, achieving the vision of equitable and accessible high-quality medical resources.
[1] Sung H, Ferlay J, Siegel RL, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2021, 71(3): 209-249.
[2] Huang YB, Dai HJ, Song FJ, et al. Preliminary effectiveness of breast cancer screening among 1.22 million Chinese females and different cancer patterns between urban and rural women[J].Sci Rep, 2016, 6(1):39459.
[3] 欧阳飞,石磊.1990—2021年中国及全球女性乳腺癌疾病负担分析和未来趋势预测[J].中国循证医学杂志,2025,25(5):497-503.
[4] 李佳蔚,尤超.乳腺影像人工智能的应用现状[J].临床放射学杂志,2024,43(7):1233-1236.
[5] 王惠,杨新武,陈飞,等.人工智能对乳腺影像辅助诊断的研究进展[J].中国医学影像学杂志,2023,31(10):1113-1116.
[6] American College of Radiology. Breast imaging reporting and data system[S].4th ed. Reston: American College of Radiology, 2003.
[7] 中国抗癌协会乳腺癌专业委员会.中国抗癌协会乳腺癌诊治指南与规范(2021年版)[J].中国癌症杂志,2021,31(10):954-1040.
[8] 赵军.钼靶X射线、超声及MRI对乳腺癌诊断的应用价值[J].现代仪器与医疗,2014,20(4):26-28.
[9] 廖运华,谢月哺,吴剑琴.彩色多普勒超声及钼靶X射线检查在乳腺癌诊断中的应用价值[J].临床医学研究与实践,2020,5(3):3-4.
[10] 莫淼,柳光宇,吕力琅,等.乳腺癌筛查研究进展[J].肿瘤,2012,32(9):748-754.
[11] 李晓,程流泉,刘梅,等.MRI、钼靶和超声对乳腺非肿块样强化病变诊断的对比研究[J].中国医学影像学杂志,2013,21(5):336-340.
[12] 汪晓红,耿道颖,顾雅佳,等.动态增强MRI鉴别乳腺良恶性病变的价值[J].放射学实践,2005,(8):662-666.
[13] 陈林,陈悦,庞芸,等.超声自动乳腺全容积扫描在乳腺占位性病变中的初步应用[J].中国医学影像技术,2011,27(7):1378-1382.
[14] 陈红,肖祎,赵巧玲.自动乳腺全容积超声成像与常规超声诊断乳腺癌价值的对比研究[J].临床超声医学杂志,2019,21(5):381-384.
[15] 潘德润,秦耿耿,陈卫国.基于深度学习的人工智能技术在乳腺癌筛查及影像诊断中的应用进展[J].国际医学放射学杂志,2019,42(1):12-15.
[16] 商亮,郭宇峰,叶伟,等.人工智能在乳腺癌诊断中应用的研究进展[J].现代肿瘤医学,2021,29(1):155-158.
[17] 高思琦,牛司华,黄剑华,等.超声人工智能在乳腺良恶性疾病鉴别诊断中的应用[J].中国超声医学杂志,2021,37(7):752-755.
[18] 龚敬,郝雯,彭卫军.人工智能技术在乳腺影像学诊断中的应用现状与展望[J].肿瘤影像学,2019,28(3):134-138.
[19] 罗晓,李安华.人工智能在乳腺癌诊治中的应用与思考[J].中华医学超声杂志(电子版),2019,16(4):247-251.
[20] Bevers TB, Helvie M, Bonaccio E, et al. Breast cancer screening and diagnosis, version 3.2018, NCCN clinical practice guidelines in oncology[J].J Natl Compr Canc Netw, 2018,16(11):1362-1389.
[21] Mann RM, Athanasiou A, Baltzer PAT, et al. Breast cancer screening in women with extremely dense breasts recommendations of the European Society of Breast Imaging (EUSOBI)[J].Eur Radiol, 2022,32(6):4036-4045.
[22] 中国超声医学工程学会仪器工程开发专业委员会,中国抗癌协会肿瘤影像专委会,中华医学会超声医学分会浅表组织与血管学组.自动乳腺容积超声技术专家共识(2022版)[J].中国超声医学杂志,2022,38(3):241-247.
[23] 徐志坚,刘炬,陈万青,等.乳腺癌机会性筛查规范路径专家共识[J].中国肿瘤,2024,33(1):6-18.
[24] U.S.Food and Drug Administration. DEN170022[EB/OL]. (2017-07-19) [2025-05-20]. https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/denovo.cfm?id=DEN170022.
[25] Federal Register. Radiology Devices; Reclassification of Medical Image Analyzers[EB/OL].(2018-06-04)[2025-05-20].https://www.federalregister.gov/documents/2018/06/04/2018-11880/radiology-devices-reclassification-of-medical-image-analyzers.
[26] U.S.Food and Drug Administration. Clinical Performance Assessment: Considerations for Computer-Assisted Detection Devices Applied to Radiology Images and Radiology Device Data in Premarket Notification (510(k)) Submissions [EB/OL].(2022-09-28) [2025-05-20].https://www.fda.gov/regulatory-information/search-fda-guidance-documents/clinical-performance-assessment-considerations-computer-assisted-detection-devices-applied-radiology.
[27] U.S.Food and Drug Administration. P200003[EB/OL].(2021-01-11)[2025-05-20].https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpma/pma.cfm?id=P200003.
[28] Seno Medical Instruments Inc. The PIONEER-01 Study of the Imagio Breast Imaging System[EB/OL].(2023-02-08)[2025-05-20].https://clinicaltrials.gov/study/NCT01943916.
[29] Otto P, Kist K, Dornbluth NC, et al. Clinical Feasibility Study of Combined Opto-Acoustic and Ultrasonic Imaging Modality Providing Coregistered Functional and Anatomical Maps of Breast Tumors. Presented at: 97th Scientific Assembly and Annual Meeting of The Radiological Society of North America[C].Chicago, IL, 2011.
[30] 原国家食品药品监督管理总局.总局关于发布医疗器械分类目录的公告(2017年第104号)[EB/OL].(2017-08-31)[2025-05-20].https://www.nmpa.gov.cn/xxgk/ggtg/ylqxggtg/ylqxqtggtg/20170904150301406.html.
[31] 国家药品监督管理局医疗器械技术审评中心.国家药监局器审中心关于发布人工智能医疗器械注册审查指导原则的通告(2022年第8号)[EB/OL].(2022-03-09)[2025-05-20].https://www.cmde.org.cn//xwdt/shpgzgg/gztg/20220309090800158.html.
[32] 国家药品监督管理局医疗器械技术审评中心.关于发布《医疗器械分类目录》子目录01、04、07、08、09、10、19、21相关产品临床评价推荐路径的通告(2022年第30号)[EB/OL].(2022-07-14) [2025-05-20].https://www.cmde.org.cn//xwdt/shpgzgg/gztg/20220714155028153.html.
[33] 国家药品监督管理局医疗器械技术审评中心.国家药监局器审中心关于发布人工智能辅助检测医疗器械(软件)临床评价注册审查指导原则的通告(2023年第38号)[EB/OL].(2023-11-07)[2025-05-20].https://www.cmde.org.cn//xwdt/shpgzgg/gztg/20231107153309174.html.
[34] 国家药品监督管理局.国家药监局关于发布创新医疗器械特别审查程序的公告(2018年第83号)[EB/OL].(2018-11-02)[2025-05-20].https://www.nmpa.gov.cn/xxgk/ggtg/ylqxggtg/ylqxqtggtg/20181105160001106.html.
[35] 国家药品监督管理局医疗器械技术审评中心.创新医疗器械特别审查申请审查结果公示(2025年第6号)[EB/OL].(2025-07-02)[2025-07-7].https://www.cmde.org.cn//xwdt/shpgzgg/cxyxgsh/20250702110914114.html.
[36] 国家药品监督管理局医疗器械技术审评中心.乳腺X射线图像辅助检测软件(CQZ2200165) [EB/OL].(2023-01-03) [2025-04-28].https://www.cmde.org.cn//xwdt/shpbg/20230117105821194.html.
[37] 国家药品监督管理局医疗器械技术审评中心.乳腺超声影像辅助检测软件(CQZ2400392)[EB/OL].(2025-02-25)[2025-05-20].https://www.cmde.org.cn//xwdt/shpbg/20250312091951142.html.
/
| 〈 |
|
〉 |