hc1 was created in the cloud in 2011, but our precision prescribing solution hc1 PGx Advisor® marks the first time our software engineers have used a cloud-native, serverless model in the design of an application itself. Being both an AWS Healthcare Competency and Advanced Technology Partner, hc1 looks to Amazon Web Services (AWS) to find tools that work best to handle the issues that are part of delivering a particular healthcare solution.
Our engineering team recognized early on in the development process that new tools were needed to meet the demands of moving our customers’ patients through the 5 phases of the hc1 PGx Advisor program:
Our software architects made an early decision to code hc1 PGx Advisor in Python instead of Java, hc1’s programming language of choice. They knew Python’s built-in libraries would make it easier to connect all the AWS services needed to build the components to support the phases of the hc1 PGX Advisor workflow. Once that decision was made, the team chose services that work well in Python, such as the AWS Serverless Application Model (SAM).
“AWS SAM makes it very easy to connect the infrastructure side to the code, so it’s all done in one project,” Senior Software Engineer Matthew Morscher explained. He was able to use AWS SAM’s shorthand notation “rather than manually writing out every little explicit detail of the infrastructure.” Not only did AWS SAM save Matt time when writing infrastructure code, it also saved time during testing because he could deploy fixes to the infrastructure and the code all at once. Spending less time on infrastructure meant he could focus on writing code for the product itself.
Amazon DynamoDB, a key-value database, was another paradigm shift for hc1, as our core hc1 CRM product is built on a relational database model. In the hc1 CRM, the emphasis is on giving users the ability to search, filter, and sort their data from all angles, so the relational model is necessary. For hc1 PGX Advisor, however, the priority is to efficiently organize data from lots of different sources in one place that the code can access as needed to complete the different tasks in each phase of the workflow, so a nonrelational model makes more sense. Amazon DynamoDB also has high performance with large volumes of data, so it adapts well to any size of patient population that our customers serve.
hc1 can adapt the hc1 PGx Advisor program according to how customers want to implement it. For example, customers might want to choose a different type of testing kit. Amazon EventBridge, a rules-based router for serverless environments, allows for this flexibility. “When one of those items changes, it would be very difficult to change normally, but EventBridge allows us to just go in and edit a rule,” Matt says.
Curiosity is an hc1 core value, and our engineers exemplify this value by always seeking out better ways to solve the complicated issues involved in transforming healthcare data into actionable insights. Mirroring the mission of hc1 of “Right Patient. Right Test. Right Prescription,” hc1 engineers are using the right platform, the right services, and the right design to deliver an innovative, personalized prescribing solution. To learn more about hc1 PGx Advisor®, request a demonstration here. If you want to help personalize healthcare, improve your HIPAA-compliant AWS skills, and work with Matt and the rest of our excellent engineering team, check out our current hc1 career opportunities.