摘要
一位新闻记者兼机器人与机器学习的新闻编辑每日新闻-关于人工智能的新研究结果已经发表。根据NewsRx记者来自中国医学科学院的新闻报道，研究表明：“这项研究旨在开发基于可解释的机器学习方法的更高性能的列线图，”本研究的资助单位包括北京自然科学基金、中国医学科学院医学创新基金、中国医学科学院医学信息中心、中国医学科学院医学科学研究所、中国医学科学院中国医学科学院学报。我们的新闻记者引用了中国医学科学院的一篇研究报道:“采用LightGBM机器学习方法结合形状相加解释(称为Explain Machine Learning,EML)筛选CT临床特征并确定所选特征的临界点,然后用Cox比例风险回归模型和Kaplan-Meier生存曲线对所选特征和临界点进行评估。”分别使用原始变量和按临界点二分法的变量构建了预测脑卒中患者30天死亡率的基于Logistic回归的诺模图。在总体和个体维度上评估了两种诺模图的性能。共纳入2982例脑卒中患者和64例临床特征，MIMIC-IV数据集的30天死亡率为23.6%。10个变量(“SOFA(脓毒症相关器官衰竭评估)”。从EML中定义“最低血糖”、“最高钠”、“年龄”、“平均spo2（血氧饱和度N）”、“最高体温”、“最高心率”、“最低BUN（血尿素氮）”、“最低WBC（白细胞）”和“Charlson合并症指数”）和有效临界点。在Cox比例Haza RDS回归模型（Cox回归）中高危险亚组的30天死亡率高于低危险亚组，对诺模图的评估发现，基于EML的诺模图不仅在NIR(净重新分类指数)、Brier评分和临床净收益方面优于传统诺模图，而且在个体维度上也有显著改善，特别是在低“最高温度”患者。
Abstract
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News-New study results on artificial intell igence have been published. According to news reporting originating from the Chi nese Academy of Medical Sciences by NewsRx correspondents, research stated, "Thi s study aimed to develop a higher performance nomogram based on explainable mach ine learning methods, and to predict the risk of death of stroke patients within 30 days based on clinical characteristics on the first day of intensive care un its (ICU) admission. Data relating to stroke patients were extracted from the Me dical Information Marketplace of the Intensive Care (MIMIC) IV and III database. " Funders for this research include Beijing Natural Science Foundation; The Cams I nnovation Fund For Medical Sciences; The Program of Chinese Academy of Medical S ciences. Our news journalists obtained a quote from the research from Chinese Academy of Medical Sciences: "The LightGBM machine learning approach together with Shapely additive explanations (termed as explain machine learning, EML) was used to sele ct clinical features and define cut-off points for the selected features. These selected features and cut-off points were then evaluated using the Cox proportio nal hazards regression model and Kaplan-Meier survival curves. Finally, logistic regression-based nomograms for predicting 30-day mortality of stroke patients w ere constructed using original variables and variables dichotomized by cut-off p oints, respectively. The performance of two nomograms were evaluated in overall and individual dimension. A total of 2982 stroke patients and 64 clinical featur es were included, and the 30-day mortality rate was 23.6% in the M IMIC-IV datasets. 10 variables (‘sofa (sepsis-related organ failure assessment)' , ‘minimum glucose', ‘maximum sodium', ‘age', ‘mean spo2 (blood oxygen saturatio n)', ‘maximum temperature', ‘maximum heart rate', ‘minimum bun (blood urea nitro gen)', ‘minimum wbc (white blood cells)' and ‘charlson comorbidity index') and r espective cut-off points were defined from the EML. In the Cox proportional haza rds regression model (Cox regression) and Kaplan-Meier survival curves, after gr ouping stroke patients according to the cut-off point of each variable, patients belonging to the high-risk subgroup were associated with higher 30-day mortalit y than those in the low-risk subgroup. The evaluation of nomograms found that th e EML-based nomogram not only outperformed the conventional nomogram in NIR (net reclassification index), brier score and clinical net benefits in overall dimen sion, but also significant improved in individual dimension especially for low ‘ maximum temperature' patients."
NETL
NSTL