人工智能辅助诊疗系统在全科门诊中的应用效果

樊爱青; 王洪兴; 李小攀; 陈广鹏; 李东; 靳涛涛; 侯进

doi:10.16766/j.cnki.issn.1674-4152.004172

人工智能辅助诊疗系统在全科门诊中的应用效果

doi: 10.16766/j.cnki.issn.1674-4152.004172

1.
上海市浦东新区孙桥社区卫生服务中心，上海 201210
2.
上海市浦东新区万祥社区卫生服务中心，上海 201313
3.
复旦大学附属中山医院健康管理中心，上海 200032
4.
复旦大学公共卫生学院，上海 200436
5.
上海市浦东新区社区卫生指导中心，上海 200129

基金项目:

上海市浦东新区优势全科医学科项目 PWYq2025-03

上海市中西医结合学会科研基金项目 2021-24

详细信息

通讯作者:
侯进，E-mail: amandafan520@163.com

中图分类号: R197.61 R499
计量
- 文章访问数: 5
- HTML全文浏览量: 1
- PDF下载量: 0
- 被引次数: 0
出版历程
- 收稿日期: 2025-04-21
- 网络出版日期: 2025-11-17

Application effectiveness of AI-assisted diagnosis and treatment systems in general practice clinics

1.
Sunqiao Communtiy Health Service Center of Pudong New District, Shanghai 201210, China

摘要

摘要: 目的探索人工智能(AI)辅助诊疗系统在基层全科门诊中的应用模式，评估其对疾病鉴别诊断效率、病史记录优化及基层医疗服务质量提升的效果，为基层医疗智能化转型提供实证支持。方法本研究采用实证研究方法，于2024年6—9月在上海浦东新区2家社区卫生服务中心部署爱达(Ada)人工智能辅助诊疗系统。通过结构化智能问诊、疾病鉴别诊断建议、病史自动生成等功能，纳入987例患者，分析其使用率、症状分布、诊断准确性及服务流程优化效果。结果 987例患者中，主要症状为腹痛112例(11.35%)、咳嗽104例(10.54%)和头痛86例(8.71%)。爱达(Ada)人工智能辅助诊疗系统诊断首位疾病(M1)以高血压46例(4.66%)、紧张型头痛45例(4.56%)和普通感冒42例(4.26%)为主，并提供了3~5个鉴别诊断建议。结论 AI系统可显著优化全科门诊初筛效率，通过结构化数据整合，Ada人工智能辅助诊疗系统显著提升基层全科门诊的疾病筛查效率和诊断规范性，建议推广“AI+家庭医生”协同模式，强化慢性病与多病种管理能力，为分级诊疗提供技术支持。
- 人工智能辅助诊疗系统 /
- 全科门诊 /
- 多病种管理 /
- 基层医疗 /
- 诊断效能 /
- 慢性病
Abstract: Objective To explore the application of an artificial intelligence (AI)-assisted diagnostic system in primary care settings and evaluate its effectiveness in improving disease differentiation efficiency, optimizing medical history documentation, and enhancing the quality of primary healthcare services, thereby providing empirical support for intelligent transformation in grassroots medical practices. Methods This empirical study deployed the Ada AI-assisted diagnostic system in two community health centers in Pudong New District, Shanghai from June to September 2024. Utilizing structured intelligent inquiry, differential diagnosis recommendations, and automated medical history generation, data from 987 patients were analyzed to assess system utilization rates, symptom distribution, diagnostic accuracy, and workflow optimization outcomes. Results Among 987 patients, the main symptoms included abdominal pain in 112 cases (11.35%), cough in 104 cases (10.54%), and headache in 86 cases (8.71%). The Ada artificial intelligence-assisted diagnosis and treatment system primarily identified hypertension (46 cases, 4.66%), tension-type headache (45 cases, 4.56%), and common cold (42 cases, 4.26%) as the top-ranked diagnoses (M1), while providing 3-5 differential diagnostic suggestions. Conclusion The AI system significantly enhances primary screening efficiency in general practice clinics. Through structured data integration, the Ada system improves disease screening efficacy and diagnostic standardization at the primary care level. We propose promoting an "AI-PCP (Primary Care Physician) synergistic model" to strengthen chronic disease and multi-morbidity management capabilities, providing technical support for hierarchical medical systems.
- AI-assisted diagnosis and treatment system /
- General practice clinic /
- Multi-disease management /
- Community healthcare /
- Diagnostic efficacy /
- Chronic disease

HTML全文

图 1 爱达(Ada)人工智能辅助诊疗系统流程

Figure 1. Flowchart of the Ada artificial intelligence-assisted diagnosis and treatment system

下载: 全尺寸图片幻灯片

图 2 使用爱达(Ada)人工智能辅助诊疗系统的患者症状分布(前100)

Figure 2. Symptom distribution of the Ada artificial intelligence-assisted diagnosis and treatment system(Top 100)

下载: 全尺寸图片幻灯片

图 3 爱达(Ada)人工智能辅助诊疗系统鉴别诊断M1分布(前100)

Figure 3. Distribution of differential diagnosis M1 in the Ada artificial intelligence-assisted diagnosis and treatment system(Top 100)

下载: 全尺寸图片幻灯片

表 1 987例患者一般资料

Table 1. General information of 987 patients

项目	例(%)
性别
男性	489(49.54)
女性	498(50.46)
年龄(岁)
< 18	45(4.56)
18~29	169(17.12)
30~39	190(19.25)
40~49	167(16.92)
50~59	175(17.73)
60~69	97(9.83)
70~79	95(9.63)
80~89	34(3.44)
90及以上	15(1.52)
高血压
是	224(22.69)
否	90(9.12)
不确定	673(68.19)
糖尿病
是	38(3.85)
否	799(80.95)
不确定	150(15.20)
吸烟
是	202(20.47)
否	56(5.67)
不确定	729(73.86)

下载: 导出CSV

参考文献(16)

[1]	支涵颖. 全国基层医疗卫生资源区域性配置差异研究[J]. 社会科学前沿, 2024, 13(5): 410-418. ZHI H Y. Research onregional allocation differences of grassroots medical and health resources in China[J]. Advances in Social Sciences, 2024, 13(5): 410-418.
[2]	田甜, 张飞, 张璇, 等. 基于ChatGPT的人工智能技术在全科医疗服务中的应用研究[J]. 中华全科医学, 2025, 23(5): 721-725. doi: 10.16766/j.cnki.issn.1674-4152.003988 TIAN T, ZHANG F, ZHANG X, et al. Research on the application of artificial intelligence techolgy based on ChatGPT in general medical services[J]. Chinese Journal of General Practice, 2025, 23(5): 721-725. doi: 10.16766/j.cnki.issn.1674-4152.003988
[3]	金红芳, 田军, 许冠华, 等. 县域医共体内慢性病患者"下转难"的调查研究[J]. 卫生经济研究, 2022, 39(7): 15-17. JIN H F, TIAN J, XU G H, et al. Research on the"difficulty referral to primary care institutions" of chronic disease patients in the county-level medical alliance[J]. Health Economics Research, 2022, 39(7): 15-17.
[4]	段晖, 刘颖, 张英楠, 等. 北京市医联体政策背景下慢病患者就医选择影响因素研究[J]. 中国卫生政策研究, 2020, 13(6): 14-20. DUAN H, LIU Y, ZHANG Y N, et al. Study on the influencing factors of the medical treatment choices of chronic disease patients under the background of Beijing integrated health care system[J]. Chinese Journal of Health Policy, 2020, 13(6): 14-20.
[5]	于洋, 李含伟. 社区医院全科医生人工智能素养的制约因素及路径调适[J]. 中国卫生事业管理, 2024, 41(9): 975-982. YU Y, LI H W. Constraints and Path Adjustment of General Practitioners' AI Literacy in Community Hospitals[J]. Chinese Health Service Management, 2024, 41(9): 975-982.
[6]	国家卫生健康委员会. 关于推进紧密型县域医疗卫生共同体建设的通知[EB/OL]. (2019-05-15)[2024-12-26]. https://www.gov.cn/zhengce/zhengceku/2019-10/08/content_5437020.htm.
[7]	国家卫生健康委员会. 卫生部办公厅关于印发《人工智能辅助诊断技术管理规范(试行)》的通知[EB/OL]. (2009-11-20)[2024-12-26]. http://www.nhc.gov.cn/zwgk/wtwj/201304/ffddea011a1c4ae498fcbfb1a125bfd5.shtml.
[8]	马海波. 过度医疗: 医疗服务管制之殇[J]. 延安大学学报(社会科学版), 2020, 42(4): 72-76. MA H B. Excessive Medical Treatment: The Induced Result of Medical Service Regulation[J]. Journal of Yan' an University (Social Sciences Edition), 2020, 42(4): 72-76.
[9]	刘延晶, 王少鹏, 徐喜卿, 等. 人工智能诊断系统在基层眼底视网膜疾病筛查领域的应用实践[J]. 眼科学报, 2023, 38(5): 405-413. LIU Y J, WANG S P, XU X Q, et al. Application practice of artificial intelligence diagnosis system in the field of primary fundus retinal disease screening[J]. Eye Science, 2023, 38(5): 405-413.
[10]	万东华, 江金女, 潘波, 等. 广东省基层医生对人工智能辅助诊疗技术的应用意愿分析[J]. 华南预防医学, 2023, 49(1): 32-36. WAN D H, JIANG J N, PAN B, et al. Application willingness of artificial intelligence‐assisted diagnosis and treatment technology among primary care physicians in Guangdong Province[J]. South China Journal of Preventive Medicine, 2023, 49(1): 32-36.
[11]	宋晨晓, 徐爱军. 高血压患者基层医疗机构就诊情况及其影响因素分析[J]. 中国公共卫生, 2018, 34(8): 1140-1142. SONG C X, XU A J. Status and influencing factors of health care seeking at grassroots medical institutions among hypertensive patients[J]. Chinese Journal of Public Health, 2018, 34(8): 1140-1142.
[12]	陈国湘, 李俊, 韦华, 等. 基于人工智能技术的全科医生培养模式探索[J]. 中华全科医学, 2021, 19(2): 167-170. doi: 10.16766/j.cnki.issn.1674-4152.001758 CHEN G X, LI J, WEI H, et al. Exploration of general practitioner training model based on artificial intelligence technology[J]. Chinese Journal of General Practice, 2021, 19(2): 167-170. doi: 10.16766/j.cnki.issn.1674-4152.001758
[13]	胡佳敏, 邱艳, 任菁菁. AI在基层医疗慢性病管理中的应用研究进展[J]. 中华全科医学, 2024, 22(3): 481-485. doi: 10.16766/j.cnki.issn.1674-4152.003431 HU J M, QIU Y, REN J J. Advances in the application of AI in chronic disease management in primary care[J]. Chinese Journal of General Practice, 2024, 22(3): 481-485. doi: 10.16766/j.cnki.issn.1674-4152.003431
[14]	吴晓航, 刘力学, 陈睛晶, 等. 白内障人工智能辅助诊断系统在社区筛查中的应用效果[J]. 眼科学报, 2021, 36(1): 4-9, 125. WU X H, LIU L X, CHEN Q J, et al. Application of artificial intelligence-assisted diagnostic system for community-based cataract screening[J]. Eye Science, 2021, 36(1): 4-9, 125.
[15]	王拥军, 刘涛, 刘子阳, 等. 人工智能在脑血管病领域中的应用[J]. 首都医科大学学报, 2025, 46(2): 177-183. WANG Y J, LIU T, LIU Z Y, et al. The applications of artificial intelligence in the field of cerebrovascular diseases[J]. Journal of Capital Medical University, 2025, 46(2): 177-183.
[16]	MENSAH E, PRINGLE C, ROBERTS G, et al. Deep learning in the management of intracranial aneurysms and cerebrovascular diseases: a review of the current literature[J]. World Neurosurg, 2022, 161: 39-45. doi: 10.1016/j.wneu.2022.02.006