Our graph-based technology makes real world classification applications possible.

From identifying patient cohorts for chronic disease management or clinical decision support to automating document workflows throughout the healthcare ecosystem, our technology can automate processes which are manual and slow.

Let the competition sell books.

Our technology goes way beyond just extracting entities such as disease and drugs or routing documents. A lot of folks can do that.

We go beyond and identify what is in the text and take action on it. In fact, we can also perform multi-label classification, with primary and secondary labeling.  To one document or a group of documents, like patient cohorts.

Whatever you can throw at it.

We can ingest data in HL7, ftp, data warehouse synchronization, via API’s, directly from an EHR or any other type of input. After ingestion, we can extract any type of data including patient names, drug therapies, tumor sizes etc.

We’re unbiased.

Our innovative system avoids classifier bias by being able to adjust for unbalanced datasets which are present in the real world.  Our technology can differentiate even minor differences between document types.

What we’re tackling:

  • Structured literature reviews by classifying raw medical text
  • Clinical documentation – identify missing or incomplete documents
  • Claims audits (see our Medical Coding for more details)

Interested?  We’d love to chat about our solutions.

  • Pathology

  • Physicians’ Notes

  • Imaging

  • Medical Literature