基于人工智能的病历质控系统构建与应用效果评估OA
Construction and application effect evaluation of artificial intelligence-based medical record quality control system
目的 分析应用人工智能病历质控系统对病案质量的影响,为进一步研究人工智能在病案管理中的应用提供参考,助力病案高质量管理.方法 广州市某三甲综合医院运用自然语言处理、机器学习等技术对接院内EMR、HIS、PACS、LIS等系统,基于知识规则库、逻辑关系库及逻辑条件库,构建人工智能病历质控系统,并于2024年7月1日正式上线人工智能病历质控系统.选取2024年4月1日—6月30日期间出院的15 422份病案作为对照组,选取2024年7月1日—9月30日期间出院的15 508份病案作为研究组,对两组病案的内涵缺陷率、缺陷类型、病案质量得分情况进行比较.住院病案质控依据国家及广东省卫生健康委员会发布的相关规定与评价标准,对病案首页、入院记录、病程记录、手术记录等书写质量进行评价,每份病案满分为100分.结果 人工智能病历质控系统上线后,病案内涵缺陷率由 57.77％下降到 43.64％(x2=617.875,P＜0.001),病案质量得分由(98.34±3.03)分提高到(99.02±2.09)分(t=-22.986,P＜0.001),其中,病案首页(x2=182.599,P＜0.001)、日常病程记录(x2=162.914,P＜0.001)、入院记录(x2=41.916,P＜0.001)、出院(死亡)记录(x2=99.568,P＜0.001)的内涵缺陷频次大幅下降.结论 病历人工智能质控系统有助于减少病案内涵缺陷,提高病案质量,助力病案高效管理.
Objective To analyze the impact of the application of the artificial intelligence(AI)medical record quality control system on medical record quality,provide a reference for further research on the application of AI in medical record man-agement,and facilitate high-quality medical record management.Methods A Grade A tertiary general hospital in Guangzhou constructed an AI medical record quality control system by integrating technologies such as natural language processing and ma-chine learning with the hospital's EMR,HIS,PACS,LIS and other systems,based on a knowledge rule base,a logical relation-ship base and a logical condition base.The system was officially launched on July 1,2024.A total of 15 422 medical records of discharged patients from April 1 to June 30 2024 were selected as the control group,and 15 508 medical records of discharged pa-tients from July 1 to September 30,2024 were selected as the study group.The connotation defect rate,defect types and medical record quality scores of the two groups were compared.The quality control of inpatient medical records was conducted in accord-ance with the relevant regulations and evaluation standards issued by the National Health Commission and the Health Commission of Guangdong Province.The writing quality was evaluated covering the medical record home page,admission records,progress notes,operation records and other documents,with a full score of 100 points for each medical record.Results After the launch of the AI medical record quality control system,the medical record connotation defect rate decreased from 57.77％to 43.64％(x2=617.875,P＜0.001),and the medical record quality score increased from(98.34±3.03)to(99.02±2.09)(t=-22.986,P＜0.001).Among them,the frequency of connotation defects in the medical record home page(x2=182.599,P＜0.001),daily progress notes(x2=162.914,P＜0.001),admission records(x2=41.916,P＜0.001),and discharge(death)records(x2=99.568,P＜0.001)decreased significantly.Conclusion The AI medical record quality control system helps reduce medical record connotation defects,improve medical record quality,and facilitate efficient medical record manage-ment.
谭华珍;邓活清;陈久华;叶润生;张文伟
广州市花都区人民医院 广东 广州 510800广州市花都区人民医院 广东 广州 510800广州市花都区人民医院 广东 广州 510800广州市花都区人民医院 广东 广州 510800广州市花都区人民医院 广东 广州 510800
医药卫生
人工智能病历质控系统病案质量内涵质控内涵缺陷项目
Artificial intelligence medical record quality control systemMedical record qualityConnotation quality controlConnotation defect items
《现代医院》 2026 (6)
942-945,4
广州市卫生健康科技一般引导项目(20251A011113)广州市花都区医疗卫生一般科研专项项目(25-HDWS-020)
评论