If you'd like to see us expand this article with more information, implementation details, pricing guidance, or code examples, let us know with GitHub Feedback!
Population Health Management is an important tool that is increasingly being used by health care providers to manage and control the escalating costs. The crux of Population Health Management is to use data to improve health outcomes. Tracking, monitoring, and bench marking are the three bastions of Population Health Management, aimed at improving clinical and health outcomes while managing and reducing cost.
In this solution, we will be leveraging the clinical and socioeconomic in-patient data generated by hospitals for population health reporting. As an example of a machine learning application with population health management, a model is utilized to predict length of hospital stay. It is geared towards hospitals and health care providers to manage and control the health care expenditure through disease prevention and management. You can learn about the data used and the length of hospital stay model in the manual deployment guide for the solution. Hospitals can use these results to optimize care management systems and focus their clinical resources on patients with more urgent need. Understanding the communities they serve through population health reporting can help hospitals transition from fee-for-service payments to value-based care while reducing costs and providing better care.
Download an SVG of this architecture.
Estimated Daily Cost: $156
Using Cortana Intelligence Suite you can put together and deploy from the ground up a Population Health Management solution by following the instructions here. To see the entire Population Health Management solution for Health care using Cortana Intelligence Suite in action without having to spin up and connect all the components manually, you can use the automated deployment option available here.
The 'Deploy' button will launch a workflow that will deploy an instance of the solution within a Resource Group in the Azure subscription you specify. The architecture diagram below shows the data flow and the end-to-end pipeline for Population Health Management Solution for Healthcare. The solution includes multiple Azure services and requires a few necessary manual steps to have a working end-to-end solution with simulated in-patient data from hospitals.
The architecture diagram above shows the solution design for Population Health Management Solution for Healthcare. The solution is composed of several Azure components that perform various tasks, viz. data ingestion, data storage, data movement, advanced analytics and visualization. Azure Event Hub is the ingestion point of raw records that will be processed in this solution. These are then pushed to Data Lake Store for storage and further processing by Azure Stream Analytics. A second Stream Analytics job sends selected data to PowerBI for near real time visualizations. Azure Data Factory orchestrates, on a schedule, the scoring of the raw events from the Azure Stream Analytics job by utilizing Azure Data Lake Analytics for processing with both USQL and R. Results of the scoring are then stored in Azure Data Lake Store and visualized using Power BI.
Post Deployment Steps
Once the solution is deployed to the subscription, you can see the various services deployed by clicking the resource group name on the final deployment screen. Alternatively you can use Azure management portal to see the resources provisioned in your resource group in your subscription. The source code of the solution as well as manual deployment instructions can be found here. The post deployment steps constitute monitoring the health of your deployment and visualizing the Population Health Report in real time as well as the results of the predictions from length of stay model. The post deployment instructions can be found here.