Semantic Technology Based Architecture

Free Your Data and Your Mind Will Follow

Clinigence has developed the first commercial (clinical) data warehouse fully based on semantic technologies (patent-pending.) This means that our solution is inherently "ready" for any type of data feed and can "make sense" of any new data type with no modification to the underlying infrastructure, or schema. Originally chosen to support future clinical data elements, such as genomic data and personalized medicine, this unique architecture offers many benefits to our clients today:

Cost-effective (EHR) Integration

Clinigence can integrate any EHR application in a matter of weeks and with no support from the EHR vendor. The only requirement is physical access to the data. If your data is hosted by the vendor or a 3rd party, we will need them to grant us access. Otherwise, our integration methodology leverages our unique semantic architecture to "hyper-deconstruct" any (EHR relational) database, extract all structured data and load them into our cloud-based (clinical) data warehouse. All this at a fraction of the cost and time typically associated with such integration projects. 

Non-binding Data Architecture

Most data warehouse technologies mandate binding the incoming data to a specific data model which may severely limit your future ability to expand your analytical scope. For example, if you strip facility information from claims data because this is not part of your model, you will no longer be able to generate reports based on this data. Modifying the model to accommodate such new data fields often becomes prohibitively expensive and customers remain bound to the vendor's model and at the vendor's mercy for any modification.

This is not the case with Clinigence's semantic architecture. Not only is our data model ready for any data type, but we deliberately opened up our data normalization tools to the point that you can do it yourself. As a matter of fact, only minimal technical skills are required. Any medical coder can easily use our data mapping tools. In the example we gave earlier, as long as facility information is available in the incoming data, your staff can find it and map it as the need arises for reports that use this information. 

Another case in point is the current switch from ICD-9 to ICD-10. For virtually all health IT vendors this switch requires a major upgrade---involved and costly for their clients. The Clinigence solution on the contrary requires absolutely no upgrade to support the new code set. The architecture has been ready for it from day one.

Data Normalization that Learns from Experience

Clinical data suffer from a great deal of variability. Even when data is structured (i.e. form-like rather than free text,) different vocabularies are used by different providers. The multitude of clinical systems and the lack of interoperability standards exacerbate this challenge. Enabling self-service data mapping may reduce the cost/effort, but it does not eliminate what some perceive as an insurmountable barrier that renders clinical data all but useless for analytical purposes. Clinigence has developed unique data normalization techniques that further automate the process and reduce the manual effort required in data mapping. Our solution leverages crowdsourcing to create a data normalization system that learns from experience and the wisdom of the crowds.

Declarative Classification Engine; Clinicians in the Driver's Seat

Finally, the Clinigence solution includes a declarative classification engine that underlies our analytic tools. This means that you can define your own metrics or reports at run time. Combine this with our semantic architecture and it becomes exponentially more powerful: the people who define the metrics need not have any knowledge of the data sources - their models or vocabularies. This means you can put clinicians in the driver's seat of defining care protocols and clinical metrics and eliminate their dependence on IT or even clinical informatics staff to generate analytic reports.