Deep learning models for ICU readmission prediction: a systematic review and meta-analysis

IF 9.3 1区医学 Q1 CRITICAL CARE MEDICINE

Critical Care Pub Date : 2025-10-17 DOI:10.1186/s13054-025-05642-x

Emanuele Koumantakis, Konstantina Remoundou, Nicoletta Colombi, Carmen Fava, Ioanna Roussaki, Alessia Visconti, Paola Berchialla

{"title":"Deep learning models for ICU readmission prediction: a systematic review and meta-analysis","authors":"Emanuele Koumantakis, Konstantina Remoundou, Nicoletta Colombi, Carmen Fava, Ioanna Roussaki, Alessia Visconti, Paola Berchialla","doi":"10.1186/s13054-025-05642-x","DOIUrl":null,"url":null,"abstract":"Intensive Care Unit (ICU) readmissions are associated with increased morbidity, mortality, and healthcare costs. Therefore, determining an appropriate timing of ICU discharge is critical. In this context, deep learning (DL) approaches have attracted significant attention. We conducted a systematic review of studies developing or validating DL models for ICU readmission prediction, published up to March 4th, 2025, and indexed in PubMed, Embase, Scopus, and Web of Science. We summarised them along multiple dimensions, including outcome and population definition, DL architecture, reproducibility, generalizability, and explainability, and provided a meta-analytic estimate of model performance. We included 24 studies encompassing 49 DL models, predominantly trained on US-based datasets, and rarely subjected to external validation. There was considerable variability across study settings, including the definition and timeframe of the ICU readmission outcome, as well as DL architecture used, alongside a substantial risk of bias. Technical reproducibility and model interpretation were rare. A meta-analysis of AUROC values from 11 studies yielded a mean of 0.78 (95% CI = 0.72–0.84), with very high heterogeneity (I2 = 99.9%). Models targeting disease-specific ICU subpopulations achieved significantly higher performance (mean AUROC = 0.92, 95% CI = 0.89–0.95, p = 0.002), and substantially lower heterogeneity (I2 = 17.1%). DL models showed promising performances in predicting ICU readmissions, but exhibited several shortcomings, including low reproducibility, over-reliance on a few US-based datasets, and limited explainability. Additionally, the high heterogeneity and risk of bias limited our ability to assess their pooled performance through meta-analysis. Taken together, our observations suggest that the quality of the evidence regarding the application of DL approaches to ICU readmission prediction is poor, thus hindering their clinical applicability.","PeriodicalId":10811,"journal":{"name":"Critical Care","volume":"34 1","pages":""},"PeriodicalIF":9.3000,"publicationDate":"2025-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Critical Care","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1186/s13054-025-05642-x","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"CRITICAL CARE MEDICINE","Score":null,"Total":0}

引用次数: 0

Abstract

Intensive Care Unit (ICU) readmissions are associated with increased morbidity, mortality, and healthcare costs. Therefore, determining an appropriate timing of ICU discharge is critical. In this context, deep learning (DL) approaches have attracted significant attention. We conducted a systematic review of studies developing or validating DL models for ICU readmission prediction, published up to March 4th, 2025, and indexed in PubMed, Embase, Scopus, and Web of Science. We summarised them along multiple dimensions, including outcome and population definition, DL architecture, reproducibility, generalizability, and explainability, and provided a meta-analytic estimate of model performance. We included 24 studies encompassing 49 DL models, predominantly trained on US-based datasets, and rarely subjected to external validation. There was considerable variability across study settings, including the definition and timeframe of the ICU readmission outcome, as well as DL architecture used, alongside a substantial risk of bias. Technical reproducibility and model interpretation were rare. A meta-analysis of AUROC values from 11 studies yielded a mean of 0.78 (95% CI = 0.72–0.84), with very high heterogeneity (I2 = 99.9%). Models targeting disease-specific ICU subpopulations achieved significantly higher performance (mean AUROC = 0.92, 95% CI = 0.89–0.95, p = 0.002), and substantially lower heterogeneity (I2 = 17.1%). DL models showed promising performances in predicting ICU readmissions, but exhibited several shortcomings, including low reproducibility, over-reliance on a few US-based datasets, and limited explainability. Additionally, the high heterogeneity and risk of bias limited our ability to assess their pooled performance through meta-analysis. Taken together, our observations suggest that the quality of the evidence regarding the application of DL approaches to ICU readmission prediction is poor, thus hindering their clinical applicability.

查看原文本刊更多论文

ICU再入院预测的深度学习模型：系统回顾和荟萃分析

重症监护室（ICU）再入院与发病率、死亡率和医疗费用的增加有关。因此，确定适当的ICU出院时间至关重要。在这种背景下，深度学习（DL）方法引起了极大的关注。我们对开发或验证ICU再入院预测DL模型的研究进行了系统回顾，这些研究发表于2025年3月4日，并在PubMed， Embase， Scopus和Web of Science中被检索。我们从多个维度对它们进行了总结，包括结果和总体定义、深度学习架构、可重复性、概括性和可解释性，并提供了模型性能的元分析估计。我们纳入了24项研究，包括49个深度学习模型，主要在美国数据集上训练，很少受到外部验证。研究设置存在相当大的差异，包括ICU再入院结果的定义和时间框架，以及使用的DL架构，以及相当大的偏倚风险。技术可重复性和模型解释是罕见的。对11项研究的AUROC值进行荟萃分析，平均为0.78 (95% CI = 0.72-0.84)，异质性非常高（I2 = 99.9%）。针对疾病特异性ICU亚群的模型取得了显著更高的性能（平均AUROC = 0.92, 95% CI = 0.89-0.95, p = 0.002），异质性显著降低（I2 = 17.1%）。深度学习模型在预测ICU再入院方面表现出良好的表现，但也存在一些缺点，包括可重复性低、过度依赖少数基于美国的数据集以及可解释性有限。此外，高异质性和偏倚风险限制了我们通过荟萃分析评估其综合表现的能力。综上所述，我们的观察结果表明，关于将DL方法应用于ICU再入院预测的证据质量较差，从而阻碍了其临床适用性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Critical Care 医学-危重病医学

CiteScore

20.60

自引率

3.30%

发文量

348

审稿时长

1.5 months

期刊介绍： Critical Care is an esteemed international medical journal that undergoes a rigorous peer-review process to maintain its high quality standards. Its primary objective is to enhance the healthcare services offered to critically ill patients. To achieve this, the journal focuses on gathering, exchanging, disseminating, and endorsing evidence-based information that is highly relevant to intensivists. By doing so, Critical Care seeks to provide a thorough and inclusive examination of the intensive care field.