Architecture Design of Healthcare Software-as-a-Service Platform for Cloud-Based Clinical Decision Support Service

Objectives: To design a cloud computing-based Healthcare Software-as-a-Service (SaaS) Platform (HSP) for delivering healthcare information services with low cost, high clinical value, and high usability. Methods: We analyzed the architecture requirements of an HSP, including the interface, business services, cloud SaaS, quality attributes, privacy and security, and multi-lingual capacity. For cloud-based SaaS services, we focused on Clinical Decision Service (CDS) content services, basic functional services, and mobile services. Microsoft's Azure cloud computing for Infrastructure-as-a-Service (IaaS) and Platform- as-a-Service (PaaS) was used. Results: The functional and software views of an HSP were designed in a layered architecture. External systems can be interfaced with the HSP using SOAP and REST/ JSON. The multi-tenancy model of the HSP was designed as a shared database, with a separate schema for each tenant through a single application, although healthcare data can be physically located on a cloud or in a hospital, depending on regulations. The CDS services were categorized into rule-based services for medications, alert registration services, and knowledge services. Conclusions: We expect that cloudbased HSPs will allow small and mid-sized hospitals, in addition to large-sized hospitals, to adopt information infrastructures and health information technology with low system operation and maintenance costs.