Comparison of machine-learning and logistic regression models to predict 30-day unplanned readmission: a development and validation study

Comparison of machine-learning and logistic regression models to predict 30-day unplanned readmission: a development and validation study
复制标题

机器学习和逻辑回归模型预测 30 天计划外再入院的比较:一项开发和验证研究

DOI:
10.1101/2023.05.06.23289569
复制
发表时间:
2023
期刊:
medRxiv
影响因子:
--
通讯作者:
Nanako Tamiya
Nanako Tamiya
中科院分区:
--
文献类型:
--
作者:
Masao Iwagami;Ryota Inokuchi;Eiryo Kawakami;Tomohide Yamada;Atsushi Goto;Toshiki Kuno;Yohei Hashimoto;Nobuaki Michihata;Tadahiro Goto;Tomohiro Shinozaki;Yu Sun;Yuta Taniguchi;Jun Komiyama;Kazuaki Uda;Toshikazu Abe;Nanako Tamiya

文献摘要

参考文献

相似文献

机器学习模型是否能优于回归模型,如逻辑回归(LR)模型,特别是当电子健康记录(EHRs)中预测变量的数量和类型增加时,这是预期的,但未知。我们的目的是比较梯度增强决策树(GBDT)、随机森林(RF)、深度神经网络(DNN)和最小绝对收缩和选择算子(LR- lasso)对意外再入院的预测性能。我们使用2015-2017年38家医院活着出院患者的电子病历进行推导,2018年进行验证,包括基本特征、诊断、手术、程序、药物代码和血液检查结果。结果是30天的意外再入院。我们创建了六种数据表模式,这些数据表具有不同数量的二元变量(≥5%或≥1%的患者或≥10例患者有),有和没有血液检测结果。对于每种数据表模式,我们使用衍生数据建立机器学习和LR模型,并使用验证数据评估每个模型的性能。在推导和验证数据集中,结果发生率分别为6.8%(23,108/339,513例出院)和6.4%(7,507/118,074例出院)。对于变量数量最少的第一个数据表(102个变量≥5%的患者具有,无血液检查结果),GBDT的c统计量最高(0.740),其次是RF (0.734), LR-LASSO(0.720)和DNN(0.664)。在变量数最多的最后一个数据表中(1543个变量≥10例患者,包括血检结果),GBDT的c统计量最高(0.764),其次是LR-LASSO(0.755)、RF(0.751)和DNN(0.720),说明GBDT与LR-LASSO的差异较小,95%置信区间重叠。综上所述,GBDT在预测非计划再入院方面普遍优于LR-LASSO,但随着变量数量的增加和血液检测结果的使用,c统计量的差异越来越小。
It is expected but unknown whether machine-learning models can outperform regression models, such as a logistic regression (LR) model, especially when the number and types of predictor variables increase in electronic health records (EHRs). We aimed to compare the predictive performance of gradient-boosted decision tree (GBDT), random forest (RF), deep neural network (DNN), and LR with the least absolute shrinkage and selection operator (LR-LASSO) for unplanned readmission. We used EHRs of patients discharged alive from 38 hospitals in 2015–2017 for derivation and in 2018 for validation, including basic characteristics, diagnosis, surgery, procedure, and drug codes, and blood-test results. The outcome was 30-day unplanned readmission. We created six patterns of data tables having different numbers of binary variables (that ≥5% or ≥1% of patients or ≥10 patients had) with and without blood-test results. For each pattern of data tables, we used the derivation data to establish the machine-learning and LR models, and used the validation data to evaluate the performance of each model. The incidence of outcome was 6.8% (23,108/339,513 discharges) and 6.4% (7,507/118,074 discharges) in the derivation and validation datasets, respectively. For the first data table with the smallest number of variables (102 variables that ≥5% of patients had, without blood-test results), the c-statistic was highest for GBDT (0.740), followed by RF (0.734), LR-LASSO (0.720), and DNN (0.664). For the last data table with the largest number of variables (1543 variables that ≥10 patients had, including blood-test results), the c-statistic was highest for GBDT (0.764), followed by LR-LASSO (0.755), RF (0.751), and DNN (0.720), suggesting that the difference between GBDT and LR-LASSO was small and their 95% confidence intervals overlapped. In conclusion, GBDT generally outperformed LR-LASSO to predict unplanned readmission, but the difference of c-statistic became smaller as the number of variables was increased and blood-test results were used.
受伤、中毒和某些其他外因后果(S00-T88)
DOI: 10.1542/9781610021074-ch19
发表时间: 2013
期刊: Pediatric ICD-10-CM 2018: A Manual for Provider-Based Coding
影响因子: --
作者:
E. Aitini
通讯作者: E. Aitini
预测研究中的验证:数据分割造成的浪费。
DOI: --
发表时间: 2018
影响因子: 7.2
作者:
E. Steyerberg
通讯作者: E. Steyerberg
使用 DPC STEM 观察铁磁准晶的局部磁畴结构
DOI: --
发表时间: 2022
期刊:
影响因子: --
作者:
Morimune-Moriya Seira;Obara Keiya;Fuseya Marika;Katanosaka Masashi;岩田健史,関岳人,石川明日香,田村隆治,幾原雄一,柴田直哉
通讯作者: 岩田健史,関岳人,石川明日香,田村隆治,幾原雄一,柴田直哉
DOI: 10.1001/jama.2011.1515
发表时间: 2011-10-19
影响因子: 120.7
作者:
Kansagara, Devan;Englander, Honora;Salanitro, Amanda;Kagen, David;Theobald, Cecelia;Freeman, Michele;Kripalani, Sunil
通讯作者: Kripalani, Sunil
DOI: 10.17632/bd63njzyjf.1
发表时间: 2021
期刊: --
影响因子: --
作者:
J. Adalsteinsson
通讯作者: J. Adalsteinsson