检验智能审核系统设计与应用  

Design and application of an intelligent inspection system

作　　者：谭伟锋[1] 戴倩莹[1] 王鹏 刘仁杰 TAN Weifeng;DAI Qianying;WANG Peng;LIU Renjie(Jiangmen Central Hospital,Jiangmen 529030,China;Chuanghui Information Technology Co.,Ltd.Guangzhou 510000,China)

机构地区：[1]江门市中心医院,广东江门529030 [2]广州创惠信息科技有限公司,广东广州510000

出　　处：《现代医院》2022年第6期911-914,共4页Modern Hospitals

基　　金：江门市医疗卫生科技计划项目(2021YL01060)。

摘　　要：目的设计一种基于历史结果演算的检验智能审核推荐指标范围系统。方法基于历史检验结果大数据演算设计算法,预先设定参数初始值,如智能审核通过率、误差率等,然后使用深度学习、动态规划等算法不断进行演算、核对、修正、迭代,从而得到一个基于大数据的算法模型,该算法可求解自动审核通过率的最佳指标范围,并统计复核误差率。结果对智能审核组与人工审核组进行统计,符合规则并自动审核归入观察组,各检验组合项目智能审核率分别是生化十项28.81%、肝胆十二项29.81%、葡萄糖测定27.06%、血常规(BCA)3.3%、C反应蛋白测定(CRP)50.49%、心肌酶组合47.39%、肝功三酶测定56.33%、生化三十项18.82%,不符合规则最后人工审核的归入对照组。对比两组审核率与审核时间,差异有统计学意义(P<0.05)。结论基于大数据的智能审核算法在检验报告审核中的应用,极大地提高了审核效率,减少工作失误,更能保证临床检验工作的医疗质量。Objective To design a system of testing and intelligent review and recommendation index range based on historical result calculation.Methods Based on the big data calculation of historical patient indicators,agree error rate,automatic review pass rate,use deep learning,dynamic programming and other algorithms to continuously perform calculations,verifications,corrections,and iterations,so as to calculate an error rate based on big data statistics and review hospitals and meet the best index range of automatic review pass rate.Results Statistics on the intelligent audit group and the manual audit group,Compliance with the rules and automatic review into observation groups,the intelligent audit rate of each inspection combination project is:Biochemical Ten Items 28.81%,Twelve items of liver and gallbladder 29.81%,Glucose determination 27.06%,Blood routine(BCA)3.3%,C-Reactive Protein Assay(CRP)50.49%,Cardiac enzyme combination 47.39%,Determination of liver function three enzymes 56.33%,thirty Biochemical Items 18.82%,Those who did not meet the final manual review of the rules were included in the control group.Comparing the review rate and review time between the two groups,the difference was statistically significant(P<0.05).Conclusion The implementation of intelligent audit improves the efficiency of inspection report review and reduces errors,ensuring the medical quality of clinical inspection work.

关 键 词：智能审核 自动化 检验报告 风险管理 

分 类 号：R193.324[医药卫生—卫生事业管理]