CHINESE JOURNAL OF MEDICINAL GUIDE >
The Application and Challenges of Artificial Intelligence in Medical Device Evaluation
Received date: 2025-09-01
Revised date: 2026-02-06
Accepted date: 2026-03-18
Online published: 2026-06-29
The safety and efficacy of medical devices are critical to public health， and evaluation constitutes a key step in ensuring the market entry of safe and effective medical devices. However， traditional evaluation models face a conflict between the rapidly increasing number of registration applications， the continuously improving level of product innovation， and the limited evaluation resources. Enhancing the quality and efficiency of the evaluation to accelerate the product launch process is a challenge for regulatory agencies and an urgent need of the industry. Artificial intelligence （AI）， with its powerful capabilities in data analysis and pattern recognition， offers a promising innovative pathway to enhance the evaluation quality and efficiency. This study begins by outlining the main development stages of AI and reviewing the research process in AI application by drug regulatory agencies in the United States， European Union， and China. It then explores potential scenarios for the AI application in medical device evaluation in China， and analyzes associated risks and implementation challenges. The study contends that with the continuous innovation in AI， its assistive application in the registration acceptance stages （format compliance review and dossier compilation）， technical evaluation stage （technical evaluation of registration application materials）， and evaluation conclusion stage （issuance of correction notices and drafting of evaluation reports） will reduce mechanical repetitive work， enabling reviewers to focus more on professional judgment and decision making， and further improve the quality and efficiency of medical device evaluation.
Yinglun Geng
,
Shijie Zhou
,
Yaohua LI
.
The Application and Challenges of Artificial
Intelligence in Medical Device Evaluation
［1］ 中国政府网.医疗器械监督管理条例（国务院令第739号）［EB/OL］.（2021-03-18）［2025-08-08］.https：//www.gov.cn/zhengce/content/2021-03/18/content_5593739.htm.
［2］ 张世庆，钱虹，李耀华.医疗器械技术审评质量管理体系建设探讨［J］.中国医疗器械杂志，2018，42（4）：276-278.
［3］ 中国政府网.中华人民共和国国民经济和社会发展第十四个五年规划和2035年远景目标纲要［EB/OL］.（2021-03-13）［2025-08-08］.http：//www.gov.cn/xinwen/2021-03/13/content_5592681.htm.
［4］ 中国政府网.国务院办公厅关于全面深化药品医疗器械监管改革促进医药产业高质量发展的意见（国办发〔2024〕53号）［EB/OL］.（2025-01-13）［2025-08-08］.https：//www.gov.cn/zhengce/content/202501/content_6996115.htm.
［5］ 徐莉，李敏，何伍.人工智能在药品开发及监管中的应用［J］.中国医药导刊，2025，27（4）：331-337.
［6］ Davis R. Expert systems： Where are we？ And where do we go from here？［J］.AI Mag， 1982，3（2）：3.
［7］ Cortes C， Vapnik V. Support-vector networks［J］.Mach Learn， 1995， 20（3）： 273-297.
［8］ Hinton GE， Osindero S， Teh YW. A fast learning algorithm for deep belief nets［J］.Neural Comput， 2006，18（7）：1527-1554.
［9］ Krizhevsky A， Sutskever I， Hinton GE. ImageNet classification with deep convolutional neural networks［J］.Commun ACM， 2017，60（6）：84-90.
［10］ He K， Zhang X， Ren S， et al. Deep residual learning for image recognition［C］//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition，2016：770-778.
［11］ Huang G， Liu Z， Van Der Maaten L， et al. Densely connected convolutional networks［C］//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition， 2017：4700-4708.
［12］ Tan M， Le Q. EfficientNet： Rethinking model scaling for convolutional neural networks［C］//International Conference on Machine Learning， 2019：6105-6114.
［13］ Bahdanau D， Cho K， Bengio Y. Neural machine translation by jointly learning to align and translate［EB/OL］.（2014-09-01）［2025-08-08］.https：//arxiv.org/abs/1409.0473.
［14］ Vaswani A， Shazeer N， Parmar N， et al. Attention is all you need［C］//Adv Neural Inf Process Syst， 2017：6000-6010.
［15］ Devlin J， Chang MW， Lee K， et al. BERT： Pre-training of deep bidirectional transformers for language understanding［C］//Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics： human language technologies， volume 1 （long and short papers）， 2019：4171-4186.
［16］ Brown T， Mann B， Ryder N， et al. Language models are few-shot learners［C］//Adv Neural Inf Process Syst 33，2020：1877-1901.
［17］ Ouyang L， Wu J， Jiang X， et al. Training language models to follow instructions with human feedback［C］//Adv Neural Inf Process Syst 35， 2022：27730-27744.
［18］ Achiam J， Adler S， Agarwal S， et al. GPT-4 technical report［EB/OL］.（2023-03-15）［2025-08-08］.https：//arxiv.org/abs/2303.08774.
［19］ Dubey A， Jauhri A， Pandey A， et al. The Llama 3 herd of models［EB/OL］.（2024-07-31）［2025-08-08］.https：//arxiv.org/abs/2407.21783.
［20］ Team G， Anil R， Borgeaud S， et al. Gemini： a family of highly capable multimodal models［EB/OL］.（2023-12-19）［2025-08-08］.https：//arxiv.org/abs/2312.11805.
［21］ Zeng A， Xu B， Wang B， et al. ChatGLM： a family of large language models from GLM-130B to GLM-4 all tools［EB/OL］.（2024-06-18）［2025-08-08］.https：//arxiv.org/abs/2406.12793.
［22］ Bai J， Bai S， Chu Y， et al. Qwen technical report［EB/OL］.（2023-09-28）［2025-08-08］.https：//arxiv.org/abs/2309.16609.
［23］ Guo D， Yang D， Zhang H， et al. Deepseek-R1： incentivizing reasoning capability in LLMs via reinforcement learning［EB/OL］.（2025-01-22）［2025-08-08］.https：//arxiv.org/abs/2501.12948.
［24］ FDA. BERTox Initiative［EB/OL］.（2025-05-09）［2025-08-08］.https：//www.fda.gov/about-fda/nctr-research-focus-areas/bertox-initiative.
［25］ FDA. FDA Announces completion of first AI-assisted scientific review pilot and aggressive agency-wide AI rollout timeline［EB/OL］.（2025-05-08）［2025-08-08］.https：//www.fda.gov/news-events/press-announcements/fda-announces-completion-first-ai-assisted-scientific-review-pilot-and-aggressive-agency-wide-ai.
［26］ EMA. Guiding principles on the use of large language models in egulatory science and for medicines regulatory activities［EB/OL］.（2024-08-29）［2025-08-08］.https：//www.ema.europa.eu/en/documents/other/guiding-principles-use-large-language-models-regulatory-science-medicines-regulatory-activities_en.pdf.
［27］ EMA. EMA qualifies first artificial intelligence tool to diagnose inflammatory liver disease （MASH） in biopsy samples［EB/OL］.（2025-03-20）［2025-08-08］.https：//www.ema.europa.eu/en/news/ema-qualifies-first-artificial-intelligence-tool-diagnose-inflammatory-liver-disease-mash-biopsy-samples.
［28］ 国家药品监督管理局.国家药监局综合司关于印发药品监管人工智能典型应用场景清单的通知（药监综函〔2024〕313号）［EB/OL］.（2024-06-18）［2025-08-08］.https：//www.nmpa.gov.cn/xxgk/fgwj/gzwj/gzwjzh/20240618144318144.html.
［29］ 国家药品监督管理局信息中心.“药监小智（测试版）”在国家药监局信息中心网站上线［EB/OL］.（2025-06-24）［2025-08-08］.https：//www.nmpaic.org.cn/yjzxyw/202506/t20250625_428167.html.
［30］ 国家药品监督管理局信息中心.江苏省基于人工智能的药品智慧审批系统［EB/OL］.（2024-05-13）［2025-11-20］.https：//www.nmpaic.org.cn/zhuanti/ypzhjgdxalzjhd/dxal/zhjgal_958/202405/t20240513_424335.html.
［31］ 湖北省药品监督管理局.湖北：以大模型推动“智能决策”［EB/OL］.（2025-07-02）［2025-11-20］.https：//mpa.hubei.gov.cn/bmdt/mtjj/202507/t20250702_5711251.shtml.
［32］ 国家药品监督管理局医疗器械技术审评中心.审评审批流程［EB/OL］.［2025-08-08］.https：//www.cmde.org.cn/sqrzc/zxfw/lcjtcn/lcjtspsp/index.html.
［33］ 国家药品监督管理局医疗器械技术审评中心.信息化赋能医疗器械审评提质增效［EB/OL］.（2024-05-17）［2025-11-25］.https：//www.cmde.org.cn//xwdt/zxyw/20240517091222142.html.
［34］ 国家药品监督管理局.国家药监局关于发布《医疗器械产品注册项目立卷审查要求》等文件的通告（2022年第40号）［EB/OL］.（2022-08-31）［2025-08-08］.https：//www.nmpa.gov.cn/xxgk/ggtg/ylqxggtg/ylqxqtggtg/20220831164520126.html.
［35］ 原国家食品药品监督管理总局.《医疗器械分类规则》（国家食品药品监督管理总局令第15号）［EB/OL］.（2015-07-14）［2025-08-08］.https：//www.nmpa.gov.cn/ylqx/ylqxfgwj/ylqxbmgzh/20150714120001554.html.
/
| 〈 |
|
〉 |