Proving the Business Case for the Internet of Things

Orion big health data to enable app ecosystem

William Payne
November 1, 2018


Health tech company Orion Health has launched a cloud based big data analytics solution for healthcare providers.  Amadeus CORE is designed to enable an ecosystem of smart apps driven by big data and machine learning.  The solution is HIPAA compliant and HITRUST certified.

Amadeus CORE is hosted on Amazon Web Services (AWS) to provide the scalability to support organisations of any size.
The solution is designed to provide a platform for data analytics, machine learning, value-based care and population health management.

Health data can be drawn upon for analysis allowing organisations the flexibility of leveraging the potential of big data now or in the future when the organisation is ready to use machine learning models to gain insights and make predictions that could improve the lives of their patients.

"Amadeus CORE helps healthcare organisations prepare for the opportunities that lie ahead in data analytics," said Ian McCrae, Founder and CEO, Orion Health. "Healthcare organisations the world over struggle with disparate and siloed information but see the value of extracting insights from their data for better decision making and more efficient management and operation. Securely storing and enabling access to these vast volumes of health data is the first step in a data journey that healthcare organisations can take to share and unleash the potential of their data."

Making the shift to big data, machine learning and artificial intelligence is something 42% of professionals believe will happen in the next three-to-five years but despite these expectations, many are struggling to come to grips with how to leverage big data and the AI capabilities that come with it, Deloitte's 2018 Global Human Capital Trends research recently found.

The analysis highlights a "readiness gap" where 72% see AI as important but only a third of those surveyed felt ready to address it. This is especially the case in healthcare where big data is something that's been spoken about for many years but the reality of the journey to achieving the benefits of what a connected data ecosystem offers has only been achieved by a few.

"Health determinants, in isolation, are interesting. In combination, they're powerful," McCrae said. "Storing and aggregating vast volumes of different data and surfacing it via data analytics and machine learning models brings healthcare organisations a step closer to realising precision medicine, which is only possible when we have the complete picture of a person's health."

Amadeus CORE has been designed as a platform to handle very large volumes of data from multiple sources and enable machine learning. Its data model is designed for rapid development and deployment. 

The solution is designed to provide an ecosystem of apps utilising RESTful APIs on pre-configured and custom data models, with secure APIs enable developers to build innovative applications on top of the shared data. Amadeus CORE utilises ANSI SQL interfaces to allow easy data access for analytics, machine learning and BI reporting use cases.

Historical information is available for re-use without having to retrieve the data a second time from the source. When data is no longer required, it can be purged from the solution. Amadeus CORE also provides off-site storage for disaster recovery purposes.

Out-of-the-box dashboards and reports provide insights such as traffic trends, message characteristics and flow patterns to improve operational efficiency. 

With patient data stored centrally in the cloud, and aggregated, the solution supports the exchange of health information. Data is queryable via APIs and a BI interface to support a range of clinical applications.

The data store provides a repository for new and non-traditional data types. Data can be stored as is in the Source Data Store with minimal pre-processing. Custom data modelling provides flexibility for storing types of data including traditional and emerging data types.