Acute critical illness is often preceded by deterioration of routinely measured clinical parameters, e.g., blood pressure and heart rate. Early clinical prediction is typically based on manually calculated screening metrics that simply weigh these parameters, such as early warning scores (EWS). The predictive performance of EWSs yields a tradeoff between sensitivity and specificity that can lead to negative outcomes for the patient. Previous work on electronic health records (EHR) trained artificial intelligence (AI) systems offers promising results with high levels of predictive performance in relation to the early, real-time prediction of acute critical illness. However, without insight into the complex decisions by such system, clinical translation is hindered. Here, we present an explainable AI early warning score (xAI-EWS) system for early detection of acute critical illness. xAI-EWS potentiates clinical translation by accompanying a prediction with information on the EHR data explaining it.

Background: The timeliness of detection of a sepsis incidence in progress is a crucial factor in the outcome for the patient. Machine learning models built from data in electronic health records can be used as an effective tool for improving this timeliness, but so far the potential for clinical implementations has been largely limited to studies in intensive care units. This study will employ a richer data set that will expand the applicability of these models beyond intensive care units. Furthermore, we will circumvent several important limitations that have been found in the literature: 1) Models are evaluated shortly before sepsis onset without considering interventions already initiated. 2) Machine learning models are built on a restricted set of clinical parameters, which are not necessarily measured in all departments. 3) Model performance is limited by current knowledge of sepsis, as feature interactions and time dependencies are hard-coded into the model. Methods: In this study, we present a model to overcome these shortcomings using a deep learning approach on a diverse multicenter data set. We used retrospective data from multiple Danish hospitals over a seven-year period. Our sepsis detection system is constructed as a combination of a convolutional neural network and a long short-term memory network. We suggest a retrospective assessment of interventions by looking at intravenous antibiotics and blood cultures preceding the prediction time. Results: Results show performance ranging from AUROC 0.856 (3 hours before sepsis onset) to AUROC 0.756 (24 hours before sepsis onset). Evaluating the clinical utility of the model, we find that a large proportion of septic patients did not receive antibiotic treatment or blood culture at the time of the sepsis prediction, and the model could therefore facilitate such interventions at an earlier point in time. Conclusion: We present a deep learning system for early detection of sepsis that is able to learn characteristics of the key factors and interactions from the raw event sequence data itself, without relying on a labor-intensive feature extraction work. Our system outperforms baseline models, such as gradient boosting, which rely on specific data elements and therefore suffer from many missing values in our dataset.

Sodium–glucose cotransporter 2 (SGLT2) inhibition reduces cardiovascular morbidity and mortality in individuals with type 2 diabetes. Beneficial effects have been attributed to increased ketogenesis, reduced cardiac fatty acid oxidation, and diminished cardiac oxygen consumption. We therefore studied whether SGLT2 inhibition altered cardiac oxidative substrate consumption, efficiency, and perfusion. Thirteen individuals with type 2 diabetes were studied after 4 weeks’ treatment with empagliflozin and placebo in a randomized, double-blind, placebo-controlled crossover study. Myocardial palmitate and glucose uptake were measured with 11C-palmitate and 18F-fluorodeoxyglucose positron emission tomography (PET)/computed tomography (CT). Oxygen consumption and myocardial external efficiency (MEE) were measured with 11C-acetate PET/CT. Resting and adenosine stress myocardial blood flow (MBF) and myocardial flow reserve (MFR) were measured using 15O-H2O PET/CT. Empagliflozin did not affect myocardial free fatty acids (FFAs) uptake but reduced myocardial glucose uptake by 57% (P < 0.001). Empagliflozin did not change myocardial oxygen consumption or MEE. Empagliflozin reduced resting MBF by 13% (P < 0.01), but did not significantly affect stress MBF or MFR. In conclusion, SGLT2 inhibition did not affect myocardial FFA uptake, but channeled myocardial substrate utilization from glucose toward other sources and reduced resting MBF. However, the observed metabolic and hemodynamic changes were modest and most likely contribute only partially to the cardioprotective effect of SGLT2 inhibition.