人工智能预测痴呆患者住院死亡率：一项多中心研究。

Artificial intelligence prediction of In-Hospital mortality in patients with dementia: A multi-center study.

机构信息

Department of Family Medicine, Chi Mei Medical Center, Tainan, Taiwan.

Department of Occupational Medicine, Chi Mei Medical Center, Tainan, Taiwan; Department of Environmental and Occupational Health, College of Medicine, National Cheng Kung University, Tainan, Taiwan; Department of Emergency Medicine, Chi Mei Medical Center, Tainan, Taiwan.

出版信息

Int J Med Inform. 2024 Nov;191:105590. doi: 10.1016/j.ijmedinf.2024.105590. Epub 2024 Aug 6.

DOI:10.1016/j.ijmedinf.2024.105590

PMID:39142178

Abstract

BACKGROUND

Prediction of mortality is very important for care planning in hospitalized patients with dementia and artificial intelligence has the potential to serve as a solution; however, this issue remains unclear. Thus, this study was conducted to elucidate this matter.

METHODS

We identified 10,573 hospitalized patients aged ≥ 45 years with dementia from three hospitals between 2010 and 2020 for this study. Utilizing 44 feature variables extracted from electronic medical records, an artificial intelligence (AI) model was constructed to predict death during hospitalization. The data was randomly separated into 70 % training set and 30 % testing set. We compared predictive accuracy among six algorithms including logistic regression, random forest, extreme gradient boosting (XGBoost), Light Gradient Boosting Machine (LightGBM), multilayer perceptron (MLP), and support vector machine (SVM). Additionally, another set of data collected in 2021 was used as the validation set to assess the performance of six algorithms.

RESULTS

The average age was 79.8 years, with females constituting 54.5 % of the sample. The in-hospital mortality rate was 6.7 %. LightGBM exhibited the highest area under the curve (0.991) for predicting mortality compared to other algorithms (XGBoost: 0.987, random forest: 0.985, logistic regression: 0.918, MLP: 0.898, SVM: 0.897). The accuracy, sensitivity, positive predictive value, and negative predictive value of LightGBM were 0.943, 0.944, 0.943, 0.542, and 0.996, respectively. Among the features in LightGBM, the three most important variables were the Glasgow Coma Scale, respiratory rate, and blood urea nitrogen. In the validation set, the area under the curve of LightGBM reached 0.753.

CONCLUSIONS

The AI prediction model demonstrates strong accuracy in predicting in-hospital mortality among patients with dementia, suggesting its potential implementation to enhance future care quality.

摘要

背景

预测死亡率对于痴呆住院患者的护理计划非常重要，人工智能有潜力成为一种解决方案；然而，这个问题仍然不清楚。因此，进行了这项研究以阐明这个问题。

方法

我们从 2010 年至 2020 年间的三家医院中确定了 10573 名年龄≥45 岁的痴呆住院患者进行这项研究。利用从电子病历中提取的 44 个特征变量，构建了一个人工智能（AI）模型来预测住院期间的死亡。数据被随机分为 70%的训练集和 30%的测试集。我们比较了包括逻辑回归、随机森林、极端梯度提升（XGBoost）、Light Gradient Boosting Machine（LightGBM）、多层感知机（MLP）和支持向量机（SVM）在内的六种算法的预测准确性。此外，还使用 2021 年收集的另一组数据作为验证集来评估六种算法的性能。

结果

平均年龄为 79.8 岁，女性占样本的 54.5%。住院死亡率为 6.7%。与其他算法相比，LightGBM 预测死亡率的曲线下面积最高（0.991）（XGBoost：0.987、随机森林：0.985、逻辑回归：0.918、MLP：0.898、SVM：0.897）。LightGBM 的准确性、灵敏度、阳性预测值和阴性预测值分别为 0.943、0.944、0.943、0.542 和 0.996。在 LightGBM 中的特征中，三个最重要的变量是格拉斯哥昏迷量表、呼吸频率和血尿素氮。在验证集中，LightGBM 的曲线下面积达到 0.753。