Solution architecture: Predict length of stay and patient flow with healthcare analytics

For the people running a healthcare facility, length of stay—the number of days from patient admission to discharge—matters. However, that number can vary across facilities and across disease conditions and specialties, even within the same healthcare system, making it harder to track patient flow and plan accordingly.

This Azure solution helps hospital administrators use the power of machine learning to predict the length of stay for in-hospital admissions, to improve capacity planning and resource utilisation. A Chief Medical Information Officer might use a predictive model to determine which facilities are overtaxed and which resources to bolster within those facilities, and a Care Line Manager might use it to determine if there will be adequate staff resources to handle the release of a patient.

Being able to predict length of stay at the time of admission helps hospitals provide higher quality care and streamline their operational workload. It also helps accurately plan for discharges, lowering other quality measures such as readmissions.

Deploy to Azure

Use the following pre-built template to deploy this architecture to Azure

Deploy to Azure
Power BI SQL Database Machine Learning

Implementation guidance

Products Documentation

SQL Server R Services

Stores the patient and hospital data. Provides training and predicted models and predicted results for consumption using R.

Power BI

Power BI provides an interactive dashboard with visualisation that uses data stored in SQL Server to drive decisions on the predictions.

Machine Learning Studio

Machine Learning helps you easily design, test, operationalise and manage predictive analytics solutions in the cloud.

Related solution architectures