EUVIMEDEUVIMED entdecken
Crossref

Machine learning for mortality prediction using nutritional and inflammatory markers in critically ill patients

A García-Grimaldo, C A Galindo-Martín, N C Rodriguez-Moguel, C M Hernández-Cárdenas, E L Nicolas-Martinez, M A Rios-Ayala, I Balam-Lara, J D Cadeza-Aguilar, M Godínez-Victoria, I A Osuna-Padilla

Clinical Nutrition ESPEN · 2026-06 · 73 · 103254

DOI: 10.1016/j.clnesp.2026.103254 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Clinical Nutrition ESPEN, 2026-06. Verfasst von A García-Grimaldo und 9 weiteren Autor:innen. Beteiligte Einrichtung: Departamento de Nutrición Clínica, Instituto Nacional de Enfermedades Respiratorias, Mexico City, Mexico; Sección de Estudios de Posgrado e Investiga…. Hauptthemen laut MEDLINE: Critical Illness, Nutritional Status, Machine Learning, Malnutrition. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 20 Literaturangaben.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

INTRODUCTION: Malnutrition and muscle loss are key determinants of outcomes in critically ill patients, yet conventional ICU mortality scores (e.g., APACHE II, SOFA) do not incorporate nutritional status. This study aimed to develop a machine learning model integrating clinical and nutritional indicators associated with malnutrition diagnosis to predict in-hospital mortality in critically ill patients with respiratory diseases. METHODS: The cohort was divided into training (70%) and test (30%) sets. An artificial neural network (NN) was trained using age, SOFA score, serum C-reactive protein (CRP), calf circumference (CC) adjusted for body mass index, and mid-upper arm circumference (MUAC), selected as nutritional and inflammatory indicators. Patients were not formally classified as malnourished; instead, individual phenotypic and etiologic indicators were used as predictors. Ten-fold cross-validation and random oversampling addressed class imbalance, and hyperparameter tuning was based on precision. Model performance was evaluated using accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and area under the receiver operating characteristic curve (AUC). RESULTS: We analyzed 226 critically ill patients. According to variable importance, the three most influential predictors were age, BMI-adjusted CC and SOFA score. The test set achieved an AUC of 0.77 (95% CI: 0.62-0.92), accuracy of 0.67, positive predictive 0.35 value and NPV of 0.89. Inclusion of nutritional and inflammatory indicators improve predictive performance. CONCLUSION: This NN-based model provides a feasible, clinically relevant tool for predicting survival in critically ill patients, integrating nutritional and inflammatory markers. Its reliance on simple, routinely collected variables enables implementation in resource-limited environments. Further studies evaluating the response to nutritional support using this model are warranted.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Critical IllnessNutritional StatusMachine LearningMalnutritionInflammationHumansFemaleBiomarkersMaleNutrition AssessmentMiddle AgedAgedC-Reactive ProteinHospital MortalityIntensive Care UnitsPredictive Learning ModelsPredictive Value of TestsBody Mass IndexROC CurveNeural Networks, Computer

Stichwörter der Autor:innen

Critical illnessHospital mortalityMalnutritionNutritional assessment

Substanzen

BiomarkersC-Reactive Protein

Autor:innen und Institutionen

  • Departamento de Nutrición Clínica, Instituto Nacional de Enfermedades Respiratorias, Mexico City, Mexico; Sección de Estudios de Posgrado e Investigación, Escuela Superior de Medicina, Instituto Politécnico Nacional. Me…
  • Servicio de Nutrición, Hospital San Angel Inn Universidad, Mexico City, Mexico.
  • Departamento de Nutrición Clínica, Instituto Nacional de Enfermedades Respiratorias, Mexico City, Mexico.
  • Dirección General, Instituto Nacional de Enfermedades Respiratorias, Mexico City, Mexico.
  • Departamento de Áreas Críticas, Instituto Nacional de Enfermedades Respiratorias, Mexico City, Mexico.
  • Sección de Estudios de Posgrado e Investigación, Escuela Superior de Medicina, Instituto Politécnico Nacional. Mexico City, Mexico.

Zitieren

APA

García-Grimaldo, A., Galindo-Martín, C. A., Rodriguez-Moguel, N. C., Hernández-Cárdenas, C. M., Nicolas-Martinez, E. L., Rios-Ayala, M. A., Balam-Lara, I., Cadeza-Aguilar, J. D., Godínez-Victoria, M., & Osuna-Padilla, I. A. (2026). Machine learning for mortality prediction using nutritional and inflammatory markers in critically ill patients. Clinical Nutrition ESPEN, 73, 103254. https://doi.org/10.1016/j.clnesp.2026.103254

Vancouver

García-Grimaldo A, Galindo-Martín CA, Rodriguez-Moguel NC, Hernández-Cárdenas CM, Nicolas-Martinez EL, Rios-Ayala MA, et al. Machine learning for mortality prediction using nutritional and inflammatory markers in critically ill patients. Clinical Nutrition ESPEN. 2026;73:103254. doi:10.1016/j.clnesp.2026.103254

BibTeX
@article{GarcaGrimaldo2026,
  author = {García-Grimaldo, A and Galindo-Martín, C A and Rodriguez-Moguel, N C and Hernández-Cárdenas, C M and Nicolas-Martinez, E L and Rios-Ayala, M A and Balam-Lara, I and Cadeza-Aguilar, J D and Godínez-Victoria, M and Osuna-Padilla, I A},
  title = {Machine learning for mortality prediction using nutritional and inflammatory markers in critically ill patients},
  journal = {Clinical Nutrition ESPEN},
  year = {2026},
  volume = {73},
  pages = {103254},
  doi = {10.1016/j.clnesp.2026.103254}
}
Kostenlos starten →