Developing Deep Learning Continuous Risk Models for Early Adverse Event Prediction in Electronic Health Records: an AKI Case Study

Abstract

Early detection of patient deterioration is key to unlocking the potential for targeted preventative care and improving patient outcomes. This protocol describes a workflow for developing deep learning continuous risk models for early prediction of future acute adverse events from electronic health records (EHR), taking the prediction of the risk of future acute kidney injury (AKI) as an exemplar. The protocol consists of 34 steps grouped into the following stages: formal problem definition, data processing, model architecture selection, risk calibration and uncertainty, and evaluating model generalisability. For the protocol to be applicable to modelling the future risk of a particular condition, the problem formulation should be clinically and physiologically plausible and there needs to be sufficient associated predictive signal in routinely collected EHR data. Prospective validation is key in evaluating whether retrospective models developed by following the proposed protocol are clinically applicable and useful.

Relevance Of Artificial Intelligence In Healthcare: An Exhaustive Review

“…Eventually, an EMR provides real-time scores that help in transferring intensive care to the patients in hospital mortality check (Escobar et al, 2016), risk of readmission, length of stay for maximum days, and discharge diagnosis (Rajkomar et al, 2018). Furthermore, with the use of EMR, future predictions of diseases could be done efficiently such as acute kidney injury (Tomašev et al, 2019). One typical case of an EMR platform that is presently being adopted is OpenMRS.…”

Let’s talk about sex: Differences in drug therapy in males and females

“…Therefore, the results may be discriminatory and be sex bias. For example, the US Department of Veterans Affairs healthcare system was used to assess the risk of acute kidney injury [328]. Female patients comprised of 6.37% of patients in the dataset and therefore algorithm performance was lower in the females, compared with the males.…”

Edge AI for Early Detection of Chronic Diseases and the Spread of Infectious Diseases: Opportunities, Challenges, and Future Directions

“…Deep learning excels at recognizing and classifying complex objects. Deep learning techniques, especially neural networks, have shown remarkable performance in early disease detection [66,67,71]. CNNs and RNNs have been used to analyze medical images and time series data for early disease detection [48].…”