Search

Recently, researchers have been applying many new machine learning techniques for predicting the risk of readmission for heart failure. Combining such techniques through ensemble schemes holds a promise to further harness predictive performance of the resulting models. To that end, we examined two ensemble schemes and applied them to a real world dataset obtained from the EMR systems for 36,245 patients from 117 hospitals across the United States over five years. Both the ensemble schemes provided similar discriminative ability (AUC: 0.70, F1-score: 0.58) that was at least equal to or better than the base models that used a single machine learning method. The clinical impact of the models using decision curve analysis showed that at a threshold predicted probability of 0.40, the ensemble models offered 20% net benefit over the treat-all and treat-none strategies. [ABSTRACT FROM AUTHOR]

One of the goals of the Precision Medicine Initiative launched in the United States in 2016 is to use innovative tools and sources in data science. We realized this goal by implementing a use case that identified patients with heart failure at Veterans Health Administration using data from the Electronic Health Records from multiple health domains between 2005 and 2013. We applied a regularized logistic regression model and predicted 30-day readmission risk for 1210 unique patients. Our validation cohort resulted in a Cstatistic of 0.84. Our top predictors of readmission were prior diagnosis of heart failure, vascular and renal diseases, and malnutrition as comorbidities, compliance with outpatient follow-up, and low socioeconomic status. This validated risk prediction scheme delivered better performance than the published models so far (C-Statistics: 0.69). It can be used to stratify patients for readmission and to aid clinicians in delivering precise health interventions. [ABSTRACT FROM AUTHOR]

Multiple orthogonal challenges around escalating costs and providing quality care plague healthcare delivery, especially in OECD countries. This research in progress paper addresses the post-operative discharge phase of the patient journey and proffers a technology-enabled model that both supports a quality care experience post-discharge but also prudent management to minimise costly unplanned readmissions and thereby subscribe to a value-based care paradigm. The chosen context is stoma patients but the solution can be easily generalised to other contexts. Next steps include the conduct of clinical trials to establish proof of concept, validity and usability. [ABSTRACT FROM AUTHOR]