Health Discovery is the Averbis healthcare text mining and machine learning platform for analyzing large amounts of patient data. It extracts diagnoses, medications, laboratory values and many other search criteria from doctors’ letters, pathology reports and other unstructured texts. The extracted information is translated into semantic codes from our medical terminology catalogue of several million terms.
Health Discovery thus enables meaningful predictions of diagnoses and courses of therapy and treatment. Patient groups can be compiled with a few mouse clicks, supported by Natural Language Processing (NLP) – whether for feasibility studies and patient recruitment for clinical trials, for diagnosis support for rare diseases, or to support medical coding specialists in medical billing.
What is new and what has been improved?
We are proud to announce our newest major release, Health Discovery Version 6.0.. This release introduces many new or enhanced features which are more user-friendly. Discover which improvements you can expect!
No limits: For those who have been looking for an easier way to use Averbis Health Discovery GUI or REST API, we have created the ability to work without restrictions in their production setting.
And in order to simplify the workflow it is now also possible to import documents into Health Discovery and get the results via bulk export. This new paradigm of being able to import documents into production scenarios gives you various new options like a simplified data import from databases, supported PDF and Microsoft Office formats and starting multiple pipeline instances.
Visualization tools and seamless integration
Say “Hello” to graphics and data visualization! We want to make analyzing your structured data as easy as possible by linking Health Discovery to popular BI tools. Connect Averbis Health Discovery to your favorite visualization tool (Tableau, Power BI, etc.) and try the seamless integration of the REST API into Python.
Always wanted to create projects, retrieve and start pipelines or analyze texts? Through the Averbis API, the client object now offers exactly just that! Using the new Averbis Python API is straightforward and connecting to the Health Discovery is easy.
New enhancements and changes
We worked on the API format and unified REST interface format and removed bugs for providing new application areas. The REST API v0 has been replaced with the more intuitive and user-friendly version 1 (API v1) with many additional API calls.
We will provide lots of new and improved annotators in Health Discovery Version 6
We provide a new annotation mapper, which allows additional externally available information to be added to existing annotations. Our discharge pipeline supplies the integrated functionality of the previous Diagnosis-related group Pipeline. In addition, pathology reports now offer improved performance with the help of the enhanced Pathology Pipeline (currently only available for German). Furthermore, we offer a new German drug tagger format to better identify a variety of medications.
In addition, we have made some improvements that are of particular interest to the tech community. We are now using the OSGi framework, which requires Apache Karaf, and Apache UIMA was migrated to v3.
If you’re interested in the more detailed specifications, they can be found in our User Manual.
If you want to dive deeper into all our improvements we recommend you to read our complete HD6 Release Notes.
For information regarding Health Discovery 6.0 or other questions please send us an email to verena.surrey(Replace this parenthesis with the @ sign)averbis.com or use our contact form.