Development, Validation, and Evaluation of a Simple Machine Learning Model to Predict Cirrhosis Mortality.
Development, Validation, and Evaluation of a Simple Machine Learning Model to Predict Cirrhosis Mortality.
复制标题
对简单的机器学习模型的开发，验证和评估，以预测肝硬化死亡率。
DOI：
10.1001/jamanetworkopen.2020.23780
复制
发表时间：
2020-11-02
影响因子：
13.8
通讯作者：
Asch SM
中科院分区：
文献类型：
--
作者：
Kanwal F;Taylor TJ;Kramer JR;Cao Y;Smith D;Gifford AL;El-Serag HB;Naik AD;Asch SM
这项队列研究比较了不同机器学习方法在预测肝硬化总体死亡率方面的情况，并利用机器学习为肝硬化患者的一种新型预后模型选择易于评分的临床变量。
一种利用从机器学习中选择的临床变量来开发传统预后模型的混合方法，能否在提高预测准确性的同时解决与可解释性相关的挑战呢？
在一项包括107939名肝硬化患者队列的预后研究中，简单的机器学习技术与更先进的集成梯度提升技术表现相当。在肝硬化死亡率模型中使用从简单机器学习中确定的临床变量，产生了一个比传统的终末期肝病模型（含钠）更具预测性的新评分。
这些研究结果表明，这种混合方法可以通过开发新的评分来改善数据驱动的风险预测，这些新评分比机器学习更透明、更具可操作性，且比传统风险评分更具预测性。
机器学习算法比传统预后模型具有更好的预测准确性，但对于临床使用来说过于复杂和不透明。
为了比较不同机器学习方法在预测肝硬化总体死亡率方面的情况，并利用机器学习为一种新型肝硬化预后模型选择易于评分的临床变量。
这项预后研究使用了一个回顾性队列，研究对象是2011年10月1日至2015年9月30日在综合的全国退伍军人事务医疗保健系统中的130家医院及附属门诊诊所就诊的成年肝硬化患者或其并发症患者。对患者随访至2018年12月31日。数据分析时间为2017年10月1日至2020年5月31日。
潜在的预测因素包括人口统计学特征；肝病病因、严重程度和并发症；医疗资源的使用；合并症情况；以及全面的实验室和药物数据。患者被随机选择用于模型开发（66.7%）和验证（33.3%）。评估了三种不同的统计和机器学习方法：梯度下降提升法、带有最小绝对收缩和选择算子（LASSO）正则化的逻辑回归以及限制选择不超过10个预测因子的带有LASSO的逻辑回归（部分路径模型）。在5折交叉验证中评估了预测因子的纳入和模型性能。最后，使用最大似然估计对在最简约（部分路径）模型中确定的预测因子进行重新拟合（肝硬化死亡率模型[CiMM]），并将其预测性能与广泛使用的终末期肝病模型（含钠）（MELD - Na）评分进行比较。
全因死亡率
在107939名肝硬化患者中（平均[标准差]年龄为62.7[9.6]岁；96.6%为男性；66.3%为白人，18.4%为非裔美国人），年死亡率在8.8%至15.3%之间。总计，32.7%的患者在索引日期后的3年内死亡，46.2%的患者在5年内死亡。预测1年死亡率的模型对于梯度下降提升法（受试者工作特征曲线下面积[AUC]，0.81；95%置信区间，0.80 - 0.82）、带有LASSO正则化的逻辑回归（AUC，0.78；95%置信区间，0.77 - 0.79）以及部分路径逻辑模型（AUC，0.78；95%置信区间，0.76 - 0.78）具有良好的区分度。所有模型都显示出良好的校准。最终的带有机器学习衍生临床变量的CiMM模型比MELD - Na评分具有显著更好的区分度，对于1年死亡率，AUC分别为0.78（95%置信区间，0.77 - 0.79）和0.67（95%置信区间，0.66 - 0.68）（DeLong z = 17.00；P <.001）。
在这项研究中，简单的机器学习技术与更先进的集成梯度提升技术表现相当。在肝硬化死亡率模型中使用从简单机器学习中确定的临床变量，产生了一个比机器学习更透明且比MELD - Na评分更具预测性的新评分。
This cohort study compares different machine learning methods in predicting overall mortality in cirrhosis and uses machine learning to select easily scored clinical variables for a novel prognostic model in patients with cirrhosis. Can a blended approach that uses clinical variables selected from machine learning to develop traditional prognostic models improve the accuracy of prediction while addressing challenges related to interpretability? In a prognostic study including a cohort of 107 939 patients with cirrhosis, simple machine learning techniques performed as well as the more advanced ensemble gradient boosting techniques. Using the clinical variables identified from simple machine learning in a cirrhosis mortality model produced a new score more predictive than the traditional Model for End Stage Liver Disease with sodium. These findings suggest that this blended approach can improve data-driven risk prognostication through the development of new scores that are both more transparent and actionable than machine learning and more predictive than traditional risk scores. Machine-learning algorithms offer better predictive accuracy than traditional prognostic models but are too complex and opaque for clinical use. To compare different machine learning methods in predicting overall mortality in cirrhosis and to use machine learning to select easily scored clinical variables for a novel cirrhosis prognostic model. This prognostic study used a retrospective cohort of adult patients with cirrhosis or its complications seen in 130 hospitals and affiliated ambulatory clinics in the integrated, national Veterans Affairs health care system from October 1, 2011, to September 30, 2015. Patients were followed up through December 31, 2018. Data were analyzed from October 1, 2017, to May 31, 2020. Potential predictors included demographic characteristics; liver disease etiology, severity, and complications; use of health care resources; comorbid conditions; and comprehensive laboratory and medication data. Patients were randomly selected for model development (66.7%) and validation (33.3%). Three different statistical and machine learning methods were evaluated: gradient descent boosting, logistic regression with least absolute shrinkage and selection operator (LASSO) regularization, and logistic regression with LASSO constrained to select no more than 10 predictors (partial pathway model). Predictor inclusion and model performance were evaluated in a 5-fold cross-validation. Last, the predictors identified in the most parsimonious (the partial path) model were refit using maximum-likelihood estimation (Cirrhosis Mortality Model [CiMM]), and its predictive performance was compared with that of the widely used Model for End Stage Liver Disease with sodium (MELD-Na) score. All-cause mortality. Of the 107 939 patients with cirrhosis (mean [SD] age, 62.7 [9.6] years; 96.6% male; 66.3% white, 18.4% African American), the annual mortality rate ranged from 8.8% to 15.3%. In total, 32.7% of patients died within 3 years, and 46.2% died within 5 years after the index date. Models predicting 1-year mortality had good discrimination for the gradient descent boosting (area under the receiver operating characteristics curve [AUC], 0.81; 95% CI, 0.80-0.82), logistic regression with LASSO regularization (AUC, 0.78; 95% CI, 0.77-0.79), and the partial path logistic model (AUC, 0.78; 95% CI, 0.76-0.78). All models showed good calibration. The final CiMM model with machine learning–derived clinical variables offered significantly better discrimination than the MELD-Na score, with AUCs of 0.78 (95% CI, 0.77-0.79) vs 0.67 (95% CI, 0.66-0.68) for 1-year mortality, respectively (DeLong z = 17.00; P < .001). In this study, simple machine learning techniques performed as well as the more advanced ensemble gradient boosting. Using the clinical variables identified from simple machine learning in a cirrhosis mortality model produced a new score more transparent than machine learning and more predictive than the MELD-Na score.
登录
查看更多内容
影响因子：
12.6
作者：
Naik, Aanand D.;Arney, Jennifer;Clark, Jack A.;Martin, Lindsey A.;Paragraph, Anne M. Walling;Stevenson, Autumn;Smith, Donna;Asch, Steven M.;Kanwal, Fasiha
通讯作者：
Kanwal, Fasiha
影响因子：
3.1
作者：
Koola, Jejo David;Ho, Samuel;Matheny, Michael E.
通讯作者：
Matheny, Michael E.
2001-2013 年美国退伍军人肝硬化和肝细胞癌负担趋势（按肝脏疾病分类
影响因子：
29.4
作者：
Beste, Lauren A.;Leipertz, Steven L.;Ioannou, George N.
通讯作者：
Ioannou, George N.
影响因子：
9.8
作者：
Bruno, Savino;Saibeni, Simone;Almasio, Piero Luigi
通讯作者：
Almasio, Piero Luigi
影响因子：
29.4
作者：
Jepsen, Peter;Vilstrup, Hendrik;Lash, Timothy L.
通讯作者：
Lash, Timothy L.