Related Experiment Video
Updated: Jul 10, 2026
Identification and Quantification of Deranged Metabolites in Critically Ill Patients Using NMR-Based Metabolomics
Published on: November 29, 2024
Machine learning-based prediction of ICU admission and mortality in Crimean-Congo hemorrhagic fever by using
Ahu Cephe1, Seyit Ali Büyüktuna2, Necla Koçhan3
1Institutional Data Management and Analytics Coordination Unit, Erciyes University, Kayseri, Türkiye.
Insights
Amino-acid profiles accurately predict Crimean-Congo Hemorrhagic Fever (CCHF) progression using machine learning. Specific amino acids like arginine and glutamic acid are key biomarkers for identifying high-risk patients and improving survival predictions.
Area of Science:
- Biochemistry
- Medical Informatics
- Infectious Diseases
Background:
- Crimean-Congo Hemorrhagic Fever (CCHF) is a severe viral illness with high mortality.
- Predicting disease progression and patient outcomes in CCHF remains challenging.
- Identifying reliable biomarkers is crucial for effective patient management and risk stratification.
Purpose of the Study:
- To evaluate the predictive power of amino-acid profiles for CCHF disease progression.
- To identify specific amino acids associated with clinical outcomes and survival in CCHF patients.
- To leverage machine learning (ML) for early risk stratification in CCHF.
Main Methods:
- Utilized amino acid profiles (32 concentrations) from 115 confirmed CCHF patients as input for ML models.
- Employed classification models (XGBOOST, LASSO) to predict intensive care unit (ICU) admission.
- Applied survival analysis models (Survival LASSO, RFSRC, Survival XGBOOST) to assess patient survival.
Main Results:
- XGBOOST and LASSO achieved high accuracy (AUC 0.958) in predicting ICU admission.
- Arginine and glutamic acid were the most significant predictors of ICU admission across models.
- Survival models demonstrated strong predictive performance, with ornithine, gamma-aminobutyric acid, and arginine being key amino acids for survival prediction.
Conclusions:
- ML models utilizing amino-acid profiles can accurately predict CCHF disease progression.
- Amino acid profiles offer valuable insights into the metabolic underpinnings of CCHF severity.
- This approach supports early risk stratification and personalized treatment strategies for CCHF patients.
Purpose:
This study aimed to evaluate the potential of amino-acid profiles to predict disease progression in patients with Crimean-Congo Hemorrhagic Fever (CCHF) and to identify metabolic biomarkers associated with clinical outcomes and survival.
Methods:
Of the 115 confirmed CCHF patients, 18 required intensive care unit (ICU) admission and 16 died. Notably, 15 of the deaths occurred among ICU patients, whereas only one death occurred outside the ICU. For each patient, 32 amino acid concentrations were used as input for machine-learning (ML) models.
Results:
Among the classification models evaluated for predicting ICU admission, XGBOOST and LASSO achieved the highest performance, each with an AUC of 0.958. Arginine and glutamic acid consistently emerged as the most predictive features across all models, followed by 1-methyl-L-histidine, tryptophan, and tyrosine, which appeared among the top variables in four of the five best-performing models. In survival analysis, the mean concordance index (and integrated Brier score) was 0.973 (0.10) for Survival LASSO, 0.971 (0.11) for RFSRC, and 0.942 (0.12) for Survival XGBOOST. In survival models, the top five amino acids contributing to predictions were ornithine, gamma-aminobutyric acid, ethanolamine, arginine, and histidine.
Conclusion:
ML models based on amino-acid profiles can accurately predict disease progression in CCHF, supporting early risk stratification and providing insights into the metabolic mechanisms underlying disease severity.