论文部分内容阅读
目前,医院感染已经成为医疗行业最突出的焦点问题。由于医院感染相当不容易控制,一旦发生,将会对患者的预后和转归造成较大的影响,不仅会加重患者的经济负担,还会给社会带来巨大的经济损失,严重时甚至会导致患者残疾或死亡。--《某大型综合医院医院感染预警预测——以血液病患者为例》针对医院感染的监测,国内相关软件公司推出了医院感染预警系统,通过对患者的医嘱信息、检查检验结果、电子病历等相关数据的抓取,结合预先设置的规则,对存在医院感染风险的患者预警。但是,预警系统只是对医院目前感染情况的反映,且准确性与预先设置的规则紧密相关。本文应用随机森林算法,通过对算法的训练,达到了对医院感染的预测目的,让医院感染科工作人员变治疗为预防性干预,避免或减少潜在感染的发生,减少患者痛苦,减轻患者费用负担,具有较大的社会效益和经济效益。
At present, nosocomial infections have become the most prominent focus in the medical industry. Because the hospital infection is not easy to control, once it occurs, it will have a greater impact on the patient’s prognosis and prognosis, which will not only aggravate the financial burden on patients, but also bring huge economic losses to the society and even lead to serious cases Patient disability or death. - “A large general hospital nosocomial infection prediction - taking patients with blood diseases as an example” For the monitoring of nosocomial infections, domestic software companies launched a hospital infection warning system, through the patient’s medical advice information, check the test results, electronic medical records And other related data to crawl, combined with pre-set rules, the risk of nosocomial infection in patients with early warning. However, the early warning system is only a reflection of the current infection in the hospital and its accuracy is closely related to the pre-set rules. In this paper, the random forest algorithm is used to predict the nosocomial infection through the training of the algorithm. The hospital infection department staff can be treated as preventive intervention to avoid or reduce the potential infection, reduce the suffering of patients and reduce the burden on the patients , With greater social and economic benefits.