Multi-threaded import, allowing multiple filter values

Multi-threading is used when importing codes. (With thanks to @18818569575)

Allowing filtering by multiple vocabularies, domains, and concept classes.

Fixed matching score so identical strings get a matching score of 1.0.

UTF-8 support

Adding UTF-8 support for international characters.

Bugfix

Fixed parent and child count (became zero in previous release).

Added export for review

Added option to export to a CSV file containing a wide range of information to support quality control. The option can be found under File --> Export for review.

Limiting index to valid concepts only

Limiting index to valid concepts only, even when the relationships with other concepts are valid.

Bugfix when accessibility mode is on

Preventing null pointer exceptions (manifesting as unresponsive mapping overview table) when Java is in accessibility mode.

See this issue.

Better searching and hierarchy info

• Switched similarity algorithm for better performance.
• Parent-child relationships are now extracted from concept_ancestor instead of concept_relationship to better cover non-condition concepts.
• Added index statistics info panel.
• Can now search the index even without having a code set loaded (for easier debugging).
• Concept information dialog now has maximize button (to maximize on second monitor), and all tables are sortable.

Showing vocab version in status bar

Minor addition: Showing vocab version in status bar (requires rebuilding of the index).

Major revision

In this version a lot has changed. For the end-user, the most important changes are:

• Ability to search for concepts by searching the names and synonyms of source concepts that map to standard concepts in the vocabulary. For example. the term 'Paracetamol' is not a name or synonym of the standard concept for Acetaminophen, even though it means the same thing. However, there are source codes with the name Paracetamol that are mapped to the Acetaminophen concepts, so we can find it that way if our source code system only refers to Paracetamol.
• Ability to investigate the context of a concept. At various places Usagi now reports the number of parents and children a concept has, which will help identify orphan concepts that are not integrated in the hierarchy. The Concept information panel provides detailed information on the parents and children, as well as source concepts that map to a concept.

On the back-end, Usagi now uses a Berkeley database to store information on concepts and their relationships. Also, Athena files are now read through a dedicated reader, which should reduce the number of read errors.

Added ability to map to Classification concepts (next to Standard concepts)

This version supports mapping to Classification concepts next to Standard concepts. Classification concepts are not allowed to be used to encode data in the Common Data Model, but can be used for example in adverse event definitions.

By default, Usagi filters to Standard concepts only, making it behave exactly as before. Uncheck Filter standard concepts to allow Classification concepts to be included as well.