Predicting Outcomes Of Continuous Renal Replacement Therapy Using Body Composition Monitoring: A Deep-learning Approach

This study evaluated the impact of repeated fluid balance monitoring on prognosis in patients with severe acute kidney injury (AKI) requiring continuous renal replacement therapy (CRRT). A multicenter retrospective study of 784 patients was conducted in eight tertiary hospitals in Korea. Results showed a marginal benefit of fluid balance for the control group compared to the event group. The performance of various machine learning methods, including recurrent neural networks, was compared to existing prognostic clinical scores.