Artificial intelligence in resuscitation: a scoping review
Background: Artificial intelligence (AI) is increasingly applied in medicine, with growing interest in its potential to improve outcomes in cardiac arrest (CA). However, the scope and characteristics of current AI applications in resuscitation remain unclear. Methods: This scoping review aims to map...
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-07-01
|
| Series: | Resuscitation Plus |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666520425001109 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850136331964907520 |
|---|---|
| author | Drieda Zace Federico Semeraro Sebastian Schnaubelt Jonathan Montomoli Giuseppe Ristagno Nino Fijačko Lorenzo Gamberini Elena G. Bignami Robert Greif Koenraad G. Monsieurs Andrea Scapigliati |
| author_facet | Drieda Zace Federico Semeraro Sebastian Schnaubelt Jonathan Montomoli Giuseppe Ristagno Nino Fijačko Lorenzo Gamberini Elena G. Bignami Robert Greif Koenraad G. Monsieurs Andrea Scapigliati |
| author_sort | Drieda Zace |
| collection | DOAJ |
| description | Background: Artificial intelligence (AI) is increasingly applied in medicine, with growing interest in its potential to improve outcomes in cardiac arrest (CA). However, the scope and characteristics of current AI applications in resuscitation remain unclear. Methods: This scoping review aims to map the existing literature on AI applications in CA and resuscitation and identify research gaps for further investigation. PRISMA-ScR framework and ILCOR guidelines were followed. A systematic literature search across PubMed, EMBASE, and Cochrane identified AI applications in resuscitation. Articles were screened and classified by AI methodology, study design, outcomes, and implementation settings. AI-assisted data extraction was manually validated for accuracy. Results: Out of 4046 records, 197 studies met inclusion criteria. Most were retrospective (90%), with only 16 prospective studies and 2 randomised controlled trials. AI was predominantly applied in prediction of CA, rhythm classification, and post-resuscitation outcome prognostication. Machine learning was the most commonly used method (50% of studies), followed by deep learning and, less frequently, natural language processing. Reported performance was generally high, with AUROC values often exceeding 0.85; however, external validation was rare and real-world implementation limited. Conclusions: While AI applications in resuscitation demonstrate encouraging performance in prediction and decision support tasks, clear evidence of improved patient outcomes or routine clinical use remains limited. Future research should focus on prospective validation, equity in data sources, explainability, and seamless integration of AI tools into clinical workflows. |
| format | Article |
| id | doaj-art-b80eeed90b3e4a69b4f2073a4f633c4b |
| institution | OA Journals |
| issn | 2666-5204 |
| language | English |
| publishDate | 2025-07-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Resuscitation Plus |
| spelling | doaj-art-b80eeed90b3e4a69b4f2073a4f633c4b2025-08-20T02:31:09ZengElsevierResuscitation Plus2666-52042025-07-012410097310.1016/j.resplu.2025.100973Artificial intelligence in resuscitation: a scoping reviewDrieda Zace0Federico Semeraro1Sebastian Schnaubelt2Jonathan Montomoli3Giuseppe Ristagno4Nino Fijačko5Lorenzo Gamberini6Elena G. Bignami7Robert Greif8Koenraad G. Monsieurs9Andrea Scapigliati10Department of Systems Medicine, University of Rome Tor Vergata, Rome, ItalyDepartment of Anaesthesia, Intensive Care and Prehospital Emergency, Maggiore Hospital Carlo Alberto Pizzardi, Bologna, ItalyPULS - Austrian Cardiac Arrest Awareness Association, Vienna, Austria; Emergency Medical Service Vienna, Austria; Department of Emergency Medicine, Medical University of Vienna, AustriaDepartment of Anaesthesia and Intensive Care, Infermi Hospital, Romagna Local Health Authority, Rimini, ItalyDepartment of Pathophysiology and Transplantation, University of Milan, Italy; Fondazione IRCCS Ca’ Granda Ospedale Maggiore Policlinico, Milan, ItalyUniversity of Maribor, Faculty of Health Sciences, SloveniaDepartment of Anaesthesia, Intensive Care and Prehospital Emergency, Maggiore Hospital Carlo Alberto Pizzardi, Bologna, ItalyAnesthesiology, Critical Care and Pain Medicine Division, Department of Medicine and Surgery, University of Parma, Parma, Italy; Corresponding author at: Anesthesiology, Critical Care and Pain Medicine Division, Department of Medicine and Surgery, University of Parma, Parma, Italy.Faculty of Medicine, University of Bern, Bern, Switzerland; Department of Surgical Science, University of Torino, Torino, ItalyDepartment of Emergency Medicine, Antwerp University Hospital and University of Antwerp, BelgiumFondazione Policlinico Universitario A. Gemelli, IRCCS, Institute of Anaesthesia and Intensive Care, Catholic University of the Sacred Heart, Rome, ItalyBackground: Artificial intelligence (AI) is increasingly applied in medicine, with growing interest in its potential to improve outcomes in cardiac arrest (CA). However, the scope and characteristics of current AI applications in resuscitation remain unclear. Methods: This scoping review aims to map the existing literature on AI applications in CA and resuscitation and identify research gaps for further investigation. PRISMA-ScR framework and ILCOR guidelines were followed. A systematic literature search across PubMed, EMBASE, and Cochrane identified AI applications in resuscitation. Articles were screened and classified by AI methodology, study design, outcomes, and implementation settings. AI-assisted data extraction was manually validated for accuracy. Results: Out of 4046 records, 197 studies met inclusion criteria. Most were retrospective (90%), with only 16 prospective studies and 2 randomised controlled trials. AI was predominantly applied in prediction of CA, rhythm classification, and post-resuscitation outcome prognostication. Machine learning was the most commonly used method (50% of studies), followed by deep learning and, less frequently, natural language processing. Reported performance was generally high, with AUROC values often exceeding 0.85; however, external validation was rare and real-world implementation limited. Conclusions: While AI applications in resuscitation demonstrate encouraging performance in prediction and decision support tasks, clear evidence of improved patient outcomes or routine clinical use remains limited. Future research should focus on prospective validation, equity in data sources, explainability, and seamless integration of AI tools into clinical workflows.http://www.sciencedirect.com/science/article/pii/S2666520425001109Cardiac arrestResuscitationArtificial intelligenceMachine learningDeep learningLarge language model |
| spellingShingle | Drieda Zace Federico Semeraro Sebastian Schnaubelt Jonathan Montomoli Giuseppe Ristagno Nino Fijačko Lorenzo Gamberini Elena G. Bignami Robert Greif Koenraad G. Monsieurs Andrea Scapigliati Artificial intelligence in resuscitation: a scoping review Resuscitation Plus Cardiac arrest Resuscitation Artificial intelligence Machine learning Deep learning Large language model |
| title | Artificial intelligence in resuscitation: a scoping review |
| title_full | Artificial intelligence in resuscitation: a scoping review |
| title_fullStr | Artificial intelligence in resuscitation: a scoping review |
| title_full_unstemmed | Artificial intelligence in resuscitation: a scoping review |
| title_short | Artificial intelligence in resuscitation: a scoping review |
| title_sort | artificial intelligence in resuscitation a scoping review |
| topic | Cardiac arrest Resuscitation Artificial intelligence Machine learning Deep learning Large language model |
| url | http://www.sciencedirect.com/science/article/pii/S2666520425001109 |
| work_keys_str_mv | AT driedazace artificialintelligenceinresuscitationascopingreview AT federicosemeraro artificialintelligenceinresuscitationascopingreview AT sebastianschnaubelt artificialintelligenceinresuscitationascopingreview AT jonathanmontomoli artificialintelligenceinresuscitationascopingreview AT giusepperistagno artificialintelligenceinresuscitationascopingreview AT ninofijacko artificialintelligenceinresuscitationascopingreview AT lorenzogamberini artificialintelligenceinresuscitationascopingreview AT elenagbignami artificialintelligenceinresuscitationascopingreview AT robertgreif artificialintelligenceinresuscitationascopingreview AT koenraadgmonsieurs artificialintelligenceinresuscitationascopingreview AT andreascapigliati artificialintelligenceinresuscitationascopingreview |