A Deep Learning Model for Dynamic Prediction of Acute Kidney Injury in Heart Failure Patientss.
Journal:
The Canadian journal of cardiology
Published Date:
Jan 16, 2026
Abstract
BACKGROUND: This study aimed to develop and validate a dynamic prediction model for acute kidney injury (AKI) in heart failure (HF) patients. METHODS: Using data from 7,636 HF patients in the MIMIC-IV v3.1 database, we constructed a Long Short-Term Memory (LSTM) model with dynamic focal loss to handle class imbalance. The study designed two prediction perspectives: short-term prediction utilized different data collection windows (6 to 72 hours) to dynamically predict the risk of AKI occurrence within subsequent specific time windows (12 to 72 hours); long-term prediction used data from specific time points after admission (12 to 72 hours) to predict the occurrence of AKI during the entire hospitalization. RESULTS: The model demonstrated robust performance across all prediction tasks (AUC range: 0.80-0.94). Analysis of prediction lead time showed that the model could provide early warnings: the median lead times for predicting AKI occurrence within 12, 24, 48, and 72 hours were 9.57, 12.89, 18.77, and 27.25 hours, respectively. Feature importance analysis revealed that urine output, Sequential Organ Failure Assessment score (SOFA score), and systolic blood pressure played dominant roles in short-term prediction, while TroponinT and history of cardiovascular surgery were more important in long-term prediction. CONCLUSIONS: The LSTM-based model proposed in this study captures dynamic physiological changes in HF patients and provides dynamic risk assessments for AKI with sufficient lead time.
Authors
Keywords
No keywords available for this article.
