Convolutional Neural Network Model for Intensive Care Unit Acute Kidney Injury Prediction

• Published in: Kidney international reports Vol. 6; no. 5; pp. 1289 - 1298
• Main Authors: Le, Sidney, Allen, Angier, Calvert, Jacob, Palevsky, Paul M., Braden, Gregory, Patel, Sharad, Pellegrini, Emily, Green-Saxena, Abigail, Hoffman, Jana, Das, Ritankar
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.05.2021; Elsevier
• Subjects: acute kidney injury; Clinical Research; convolutional neural net; electronic health record data; machine learning; prediction; serum creatinine; prediction; acute kidney injury; electronic health record data; machine learning; convolutional neural net; serum creatinine
• ISSN: 2468-0249; 2468-0249
• DOI: 10.1016/j.ekir.2021.02.031

Acute kidney injury (AKI) is common among hospitalized patients and has a significant impact on morbidity and mortality. Although early prediction of AKI has the potential to reduce adverse patient outcomes, it remains a difficult condition to predict and diagnose. The purpose of this study was to evaluate the ability of a machine learning algorithm to predict for AKI as defined by Kidney Disease: Improving Global Outcomes (KDIGO) stage 2 or 3 up to 48 hours in advance of onset using convolutional neural networks (CNNs) and patient electronic health record (EHR) data. A CNN prediction system was developed to use EHR data gathered during patients’ stays to predict AKI up to 48 hours before onset. A total of 12,347 patient encounters were retrospectively analyzed from the Medical Information Mart for Intensive Care III (MIMIC-III) database. An XGBoost AKI prediction model and the sequential organ failure assessment (SOFA) scoring system were used as comparators. The outcome was AKI onset. The model was trained on routinely collected patient EHR data. Measurements included area under the receiver operating characteristic (AUROC) curve, positive predictive value (PPV), and a battery of additional performance metrics for advance prediction of AKI onset. On a hold-out test set, the algorithm attained an AUROC of 0.86 and PPV of 0.24, relative to a cohort AKI prevalence of 7.62%, for long-horizon AKI prediction at a 48-hour window before onset. A CNN machine learning-based AKI prediction model outperforms XGBoost and the SOFA scoring system, revealing superior performance in predicting AKI 48 hours before onset, without reliance on serum creatinine (SCr) measurements. [Display omitted]