Objective:To design a hospital pre-triage system based on the Spark big data platform and the improved Adaboost algorithm,and to pre-triage patients in the hospital in advance and accelerate the process of medical treatment.Methods:Based on the Spark big data platform,the basic data from patients entering the hospital for the first time was collected in real time,and the blockchain technology was applied to the whole process of data collection,storage and transmission,and the data was analyzed by the improved the Adaboost algorithm.The outpatient data of The 940th Hospital of the PLA Joint Logistics Support Force in the 10 years from 2011 to 2020 were used as the dataset to quickly identify and guide patients to seek medical treatment in the hospital.The application effect of the hospital pre-triage system based on the Spark big data platform and the improved Adaboost algorithm was analyzed.Results:When the custom limit weight threshold of the improved Adaboost algorithm was set to 0.52,the algorithm accuracy reached a peak of 95.56%,and the accuracy of pre-test triage was 4.24%higher than that of the traditional Adaboost algorithm.The average waiting time for patients was shortened from 0.8 h before the triage to 0.5 h,and the average consultation time for patients was shortened from 6 min before the triage to 4.8 min.Conclusion:The hospital pre-triage system based on the big data platform and the improved Adaboost algorithm can pre-triage patients before diagnosis in advance,improve the efficiency and accuracy of the triage,and relieve the hospital visiting pressure.
关键词
预分检/实时采集/Spark大数据平台/改进Adaboost算法
Key words
Pre-triage/Real-time collection/Spark big data platform/Improved Adaboost algorithm
维普
万方数据