摘要
《机器人与机器学习每日新闻-神经系统疾病和条件的新研究-谵妄》是一篇报道的主题。根据NewsRx记者来自德国乌尔姆的新闻报道，研究表明：“术后谵妄(POD)是老年患者常见的并发症，发生率为14-56%。为了实施预防措施，有必要识别有POD风险的患者。”这项研究的财政支持来自德国联邦教育和研究部。我们的新闻编辑引用了Ulm大学的研究，“在本研究中，我们的目标是与PAWEL(择期手术中的患者安全、COS T效应和生活质量)项目密切合作，开发一个用于老年患者POD预测的机器学习(ML)模型。该模型是在PAWEL研究的878名患者(无干预，年龄70岁，70岁，70岁)的数据集上训练的。采用混淆评估法和图表回顾法确定POD的存在，根据领域知识、伦理考虑和递归特征消除选择15个特征，训练Logistic回归和线性支持向量机(SVM)，并使用Receiver Operator特征(ROC)进行评估。选择的特征为美国麻醉学家学会评分、多发病率、切口缝合时间、切口缝合估计肾小球滤过率、多药治疗、心肺转流术的使用、蒙特利尔认知评估子评分“记忆”、“定向”和“语言流畅性”、发作前痴呆、临床脆弱量表、年龄、近期跌倒、术后隔离和术前苯二氮卓类药物。线性SVM表现最好，训练集的ROC曲线下面积为0.82[95%CI 0.78-0.85]。在试验集中为0.81[95%CI 0.71-0.88]，在跨中心验证中为0.76[95%CI 0.71-0.79]。为POD预测提供了一个临床上有用和可解释的ML模型。”据新闻编辑称，该研究得出结论：“该模式将被部署到老年共同管理和人工智能项目的支持性手术中。”
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News – New research on Nervous System Diseases and Condi tions - Delirium is the subject of a report. According to news reporting origina ting from Ulm, Germany, by NewsRx correspondents, research stated, “Post-operati ve delirium (POD) is a common complication in older patients, with an incidence of 14-56 %. To implement preventative procedures, it is necessary to identify patients at risk for POD.” Financial support for this research came from German Federal Ministry of Educati on and Research. Our news editors obtained a quote from the research from Ulm University, “In the present study, we aimed to develop a machine learning (ML) model for POD predic tion in older patients, in close cooperation with the PAWEL (patient safety, cos t-effectiveness and quality of life in elective surgery) project. The model was trained on the PAWEL study’s dataset of 878 patients (no intervention, age 70, 2 09 with POD). Presence of POD was determined by the Confusion Assessment Method and a chart review. We selected 15 features based on domain knowledge, ethical c onsiderations and a recursive feature elimination. A logistic regression and a l inear support vector machine (SVM) were trained, and evaluated using receiver op erator characteristics (ROC). The selected features were American Society of Ane sthesiologists score, multimorbidity, cut-to-suture time, estimated glomerular f iltration rate, polypharmacy, use of cardio-pulmonary bypass, the Montreal cogni tive assessment subscores ‘memory’, ‘orientation’ and ‘verbal fluency’, pre-exis ting dementia, clinical frailty scale, age, recent falls, post-operative isolati on and pre-operative benzodiazepines. The linear SVM performed best, with an ROC area under the curve of 0.82 [95% CI 0.78-0.85 ] in the training set, 0.81 [95% CI 0.71-0.88] in the test set and 0.76 [95 % CI 0.71-0.79] in a cross-centre validation. W e present a clinically useful and explainable ML model for POD prediction.” According to the news editors, the research concluded: “The model will be deploy ed in the Supporting SURgery with GEriatric Co-Management and AI project.”
NETL
NSTL