EUVIMEDOpdag EUVIMED
Open access

Revising Terminology and Phenotype Definitions for AI-Based Decision Support in Fluid Therapy: Call for Consensus

Audrius Andrijauskas, Tomas Jovaisa, Christer Svensen, Povilas Andrijauskas, Axel Kerroum, Vaidotas Marozas, Darius Dilijonas

Journal of Medical Internet Research · 2026-11-09 · 28 · e103138-e103138

DOI: 10.2196/103138 ↗

At a glance

Journal article in Journal of Medical Internet Research, published online on 2026-11-09. Written by Audrius Andrijauskas and 6 co-authors. Institution: Clinic of Anaesthesiology and Intensive Care, Institute of Clinical Medicine, Faculty of Medicine, Vilnius University, Vilnius, Lithuania. Major MEDLINE topics: Artificial Intelligence, Decision Support Systems, Clinical, Fluid Therapy, Terminology as Topic. 49 references. The full text is freely available in PubMed Central.

AbstractSubjectsBibliografiske oplysningerCiteRelated studies

Abstract (original language)

Intravenous fluid therapy is a cornerstone of optimizing body hydration, circulation, and perfusion, yet its rational implementation is limited by inconsistent clinical terminology. As fluid therapy moves toward AI-empowered clinical decision support systems (AI-CDSS), there is a critical need for high-quality, verified training data. The reliability of these systems is currently threatened by misleading data resulting from ambiguous definitions and the interchangeable use of nonsynonymous terms, such as “dehydration” and “hypovolemia.” This multidisciplinary panel-of-experts opinion article intends to start a discussion by identifying the key drawbacks in current terminology and proposes solutions, including a shift toward adequacy-based phenotypes and categorizing blood and water volumes as “insufficient,” “optimal,” or “excessive” rather than relying on context-sensitive quantitative measures such as “low,” “normal,” or “high.” Illustrative scenarios are used to show how inconsistent terminology and static quantitative labels could mislead an AI-CDSS. The proposed phenotype matrix is designed to provide a structured, systematic description of fluid requirements for specific clinical objectives that can be integrated into clinical algorithms and machine learning databases. An example of a fluid requirement assessment algorithm is used for exploring its applicability in AI training. The proposed terminology and phenotype matrix are provisional: they represent the opinion of a multidisciplinary expert panel and have not undergone formal consensus, clinical validation, or testing within an AI system. Broad consensus-building initiatives will follow.

Subjects

MeSH terms (MEDLINE)

Artificial IntelligenceDecision Support Systems, ClinicalFluid TherapyTerminology as TopicHumansConsensusPhenotype

Author keywords

AIclinical decision support systemsfluid therapyhemodynamic monitoring

Authors and institutions

  • Clinic of Anaesthesiology and Intensive Care, Institute of Clinical Medicine, Faculty of Medicine, Vilnius University, Vilnius, Lithuania.
  • Department of Clinical Science and Education, Section of Anesthesiology and Intensive Care, Karolinska Institutet Södersjukhuset, Stokholm, Sweden.
  • Department of Anaesthesiology and Intensive Care, Vilnius University Hospital Santros Klinikos, Vilnius, Lithuania.
  • Department of Anesthesiology and Intensive Care, Lausanne University Hospital (CHUV), Rue du Bugnon, 46, Lausanne, Vaud, 1005, Switzerland, 41 213141111.
  • Biomedical Engineering Institute, Kaunas University of Technology, Kaunas, Lithuania.
  • Electronics Engineering Department, Faculty of Electrical and Electronics Engineering, Kaunas University of Technology, Kaunas, Lithuania.

Cite

APA

Andrijauskas, A., Jovaisa, T., Svensen, C., Andrijauskas, P., Kerroum, A., Marozas, V., & Dilijonas, D. (2026). Revising Terminology and Phenotype Definitions for AI-Based Decision Support in Fluid Therapy: Call for Consensus. Journal of Medical Internet Research, 28, e103138-e103138. https://doi.org/10.2196/103138

Vancouver

Andrijauskas A, Jovaisa T, Svensen C, Andrijauskas P, Kerroum A, Marozas V, et al. Revising Terminology and Phenotype Definitions for AI-Based Decision Support in Fluid Therapy: Call for Consensus. Journal of Medical Internet Research. 2026;28:e103138-e103138. doi:10.2196/103138

BibTeX
@article{Andrijauskas2026,
  author = {Andrijauskas, Audrius and Jovaisa, Tomas and Svensen, Christer and Andrijauskas, Povilas and Kerroum, Axel and Marozas, Vaidotas and Dilijonas, Darius},
  title = {Revising Terminology and Phenotype Definitions for AI-Based Decision Support in Fluid Therapy: Call for Consensus},
  journal = {Journal of Medical Internet Research},
  year = {2026},
  volume = {28},
  pages = {e103138--e103138},
  doi = {10.2196/103138}
}
Start for free →