Transforming Clinical Data into Actionable Prognosis Models: Machine-Learning Framework and Field-Deployable App to Predict Outcome of Ebola Patients

Background: Assessment of the response to the 2014–15 Ebola outbreak indicates the need for innovations in data collection, sharing, and use to improve case detection and treatment. Here we introduce a Machine Learning pipeline for Ebola Virus Disease (EVD) prognosis prediction, which packages the best models into a mobile app to be available in clinical care settings. The pipeline was trained on a public EVD clinical dataset, from 106 patients in Sierra Leone. Methods/Principal Findings: We used a new tool for exploratory analysis, Mirador, to identify the most informative clinical factors that correlate with EVD outcome. The small sample size and high prevalence of missing records were significant challenges. We applied multiple imputation and bootstrap sampling to address missing data and quantify overfitting. We trained several predictors over all combinations of covariates, which resulted in an ensemble of predictors, with and without viral load information, with an area under the receiver operator characteristic curve of 0.8 or more, after correcting for optimistic bias. We ranked the predictors by their F1-score, and those above a set threshold were compiled into a mobile app, Ebola CARE (Computational Assignment of Risk Estimates). Conclusions/Significance: This method demonstrates how to address small sample sizes and missing data, while creating predictive models that can be readily deployed to assist treatment in future outbreaks of EVD and other infectious diseases. By generating an ensemble of predictors instead of relying on a single model, we are able to handle situations where patient data is partially available. The prognosis app can be updated as new data become available, and we made all the computational protocols fully documented and open-sourced to encourage timely data sharing, independent validation, and development of better prediction models in outbreak response. Author Summary: We introduce a machine-learning framework and field-deployable app to predict outcome of Ebola patients from their initial clinical symptoms. Recent work from other authors also points out to the clinical factors that can be used to better understand patient prognosis, but there is currently no predictive model that can be deployed in the field to assist health care workers. Mobile apps for clinical diagnosis and prognosis allow using more complex models than the scoring protocols that have been traditionally favored by clinicians, such as Apgar and MTS. Furthermore, the WHO Ebola Interim Assessment Panel has recently concluded that innovative tools for data collection, reporting, and monitoring are needed for better response in future outbreaks. However, incomplete clinical data will continue to be a serious problem until more robust and standardized data collection systems are in place. Our app demonstrates how systematic data collection could lead to actionable knowledge, which in turn would trigger more and better collection, further improving the prognosis models and the app, essentially creating a virtuous cycle.