Artificial intelligence for clinical decision support for monitoring patients in cardiovascular ICUs: A systematic review

被引:30
作者
Moazemi, Sobhan [1 ,2 ]
Vahdati, Sahar [3 ]
Li, Jason [3 ]
Kalkhoff, Sebastian [1 ,2 ]
Castano, Luis J. V. [1 ,2 ]
Dewitz, Bastian [1 ,2 ]
Bibo, Roman [1 ,2 ]
Sabouniaghdam, Parisa [4 ]
Tootooni, Mohammad S. [5 ]
Bundschuh, Ralph A. [6 ]
Lichtenberg, Artur [1 ,2 ]
Aubin, Hug [1 ,2 ]
Schmid, Falko [1 ,2 ]
机构
[1] Digital Hlth Lab Dusseldorf, Med Fac, Dept Cardiovasc Surg, Dusseldorf, Germany
[2] Univ Hosp Dusseldorf, Dusseldorf, Germany
[3] Inst Appl Informat InfAI, Dresden, Germany
[4] Heinrich Hertz Europakolleg, Dept Comp Sci, Bonn, Germany
[5] Loyola Univ Chicago, Dept Hlth Informat & Data Sci, Chicago, IL USA
[6] Univ Augsburg, Med Fac, Nucl Med, Augsburg, Germany
关键词
artificial intelligence (AI); machine learning (ML); clinical decision support (CDS); cardiovascular; intensive care unit (ICU); patient monitoring; explainable AI (XAI); MULTICENTER; PREDICTION; VALIDATION; SELECTION; FAILURE; MODEL;
D O I
10.3389/fmed.2023.1109411
中图分类号
R5 [内科学];
学科分类号
1002 ; 100201 ;
摘要
BackgroundArtificial intelligence (AI) and machine learning (ML) models continue to evolve the clinical decision support systems (CDSS). However, challenges arise when it comes to the integration of AI/ML into clinical scenarios. In this systematic review, we followed the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA), the population, intervention, comparator, outcome, and study design (PICOS), and the medical AI life cycle guidelines to investigate studies and tools which address AI/ML-based approaches towards clinical decision support (CDS) for monitoring cardiovascular patients in intensive care units (ICUs). We further discuss recent advances, pitfalls, and future perspectives towards effective integration of AI into routine practices as were identified and elaborated over an extensive selection process for state-of-the-art manuscripts. MethodsStudies with available English full text from PubMed and Google Scholar in the period from January 2018 to August 2022 were considered. The manuscripts were fetched through a combination of the search keywords including AI, ML, reinforcement learning (RL), deep learning, clinical decision support, and cardiovascular critical care and patients monitoring. The manuscripts were analyzed and filtered based on qualitative and quantitative criteria such as target population, proper study design, cross-validation, and risk of bias. ResultsMore than 100 queries over two medical search engines and subjective literature research were developed which identified 89 studies. After extensive assessments of the studies both technically and medically, 21 studies were selected for the final qualitative assessment. DiscussionClinical time series and electronic health records (EHR) data were the most common input modalities, while methods such as gradient boosting, recurrent neural networks (RNNs) and RL were mostly used for the analysis. Seventy-five percent of the selected papers lacked validation against external datasets highlighting the generalizability issue. Also, interpretability of the AI decisions was identified as a central issue towards effective integration of AI in healthcare.
引用
收藏
页数:18
相关论文
共 94 条
[71]  
Roller R., 2022, ARXIV
[72]   Multiparameter Intelligent Monitoring in Intensive Care II: A public-access intensive care unit database [J].
Saeed, Mohammed ;
Villarroel, Mauricio ;
Reisner, Andrew T. ;
Clifford, Gari ;
Lehman, Li-Wei ;
Moody, George ;
Heldt, Thomas ;
Kyaw, Tin H. ;
Moody, Benjamin ;
Mark, Roger G. .
CRITICAL CARE MEDICINE, 2011, 39 (05) :952-960
[73]  
Sattar Y., 2022, StatPearls
[74]   Predicting sepsis with a recurrent neural network using the MIMIC III database [J].
Scherpf, Matthieu ;
Graesser, Felix ;
Malberg, Hagen ;
Zaunseder, Sebastian .
COMPUTERS IN BIOLOGY AND MEDICINE, 2019, 113
[75]  
Shaban-Nejad A., 2021, Explainable AI in healthcare and medicine: building a culture of transparency and accountability, P1, DOI DOI 10.1007/978-3-030-53352-6_1
[77]   From Blackbox to Explainable AI in Healthcare: Existing Tools and Case Studies [J].
Srinivasu, Parvathaneni Naga ;
Sandhya, N. ;
Jhaveri, Rutvij H. ;
Raut, Roshani .
MOBILE INFORMATION SYSTEMS, 2022, 2022
[78]   An overview of clinical decision support systems: benefits, risks, and strategies for success [J].
Sutton, Reed T. ;
Pincock, David ;
Baumgart, Daniel C. ;
Sadowski, Daniel C. ;
Fedorak, Richard N. ;
Kroeker, Karen, I .
NPJ DIGITAL MEDICINE, 2020, 3 (01)
[79]  
Sutton RS., MACH LEARN, V173
[80]   Application of Machine Learning in Intensive Care Unit (ICU) Settings Using MIMIC Dataset: Systematic Review [J].
Syed, Mahanazuddin ;
Syed, Shorabuddin ;
Sexton, Kevin ;
Syeda, Hafsa Bareen ;
Garza, Maryam ;
Zozus, Meredith ;
Syed, Farhanuddin ;
Begum, Salma ;
Syed, Abdullah Usama ;
Sanford, Joseph ;
Prior, Fred .
INFORMATICS-BASEL, 2021, 8 (01)