Key result
A shallow neural network model outperformed Adaboost, XGBoost, and SVM in predicting post-filter ionized calcium levels during regional citrate anticoagulation, achieving an F1-score of 90.77%.
Why the study?
Regional citrate anticoagulation during continuous renal replacement therapy requires improved patient safety through early warning of citric acid overdose and dose adjustment using computer-assisted monitoring.
Can machine learning models provide early warning of citric acid overdose and guide adjustment of regional citrate anticoagulation in ICU patients?
Cohort (n=393)
No
Can machine learning models provide early warning of citric acid overdose and guide adjustment of regional citrate anticoagulation in ICU patients?
Machine learning models, particularly shallow neural networks, can effectively predict citric acid overdose and assist in adjusting regional citrate anticoagulation during continuous renal replacement therapy.
Answers are grounded in this paper and cite their sources.
May support ML monitoring of citrate anticoagulation in CRRT; leaves open prospective validation before clinical use.
BACKGROUND: Regional citrate anticoagulation (RCA) is an important local anticoagulation method during bedside continuous renal replacement therapy. To improve patient safety and achieve computer assisted dose monitoring and control, we took intensive care units patients into cohort and aiming at developing a data-driven machine learning model to give early warning of citric acid overdose and provide adjustment suggestions on citrate pumping rate and 10% calcium gluconate input rate for RCA treatment. METHODS: solvent, replacement fluid solvent, body temperature value, and replacement fluid PH value as clinical features, models attempted to classify patients who received regional citrate anticoagulation into correct outcome category. Four models, Adaboost, XGBoost, support vector machine (SVM) and shallow neural network, were compared on the performance of predicting outcomes. Prediction results were evaluated using accuracy, precision, recall and F1-score. RESULTS: For classifying patients at the early stages of citric acid treatment, the accuracy of neutral networks model is higher than Adaboost, XGBoost and SVM, the F1-score of shallow neutral networks (90.77%) is overall outperformed than other models (88.40%, 82.17% and 88.96% for Adaboost, XGBoost and SVM). Extended experiment and validation were further conducted using the MIMIC-III database, the F1-scores for shallow neutral networks, Adaboost, XGBoost and SVM are 80.00%, 80.46%, 80.37% and 78.90%, the AUCs are 0.8638, 0.8086, 0.8466 and 0.7919 respectively. CONCLUSION: The results of this study demonstrated the feasibility and performance of machine learning methods for monitoring and adjusting local regional citrate anticoagulation, and further provide decision-making recommendations to clinicians point-of-care.
No takes yet. Share an insight, caveat, or question.
Chen et al. (2021) conducted a cohort in Critically ill patients receiving regional citrate anticoagulation during continuous renal replacement therapy (n=393). Shallow neural network model vs. Adaboost, XGBoost, and SVM models was evaluated on F1-score for classifying patients at the early stages of citric acid treatment. A shallow neural network model outperformed Adaboost, XGBoost, and SVM in predicting post-filter ionized calcium levels during regional citrate anticoagulation, achieving an F1-score of 90.77%.
Synapse has enriched 10 related papers. Consider them for comparative context: