If you are using the Fast dictionary lookup component {1], and using the database that is available to download from sourceforge, then the TUI and CUI codes are coming from that database, which was created from the UMLS Metathesaurus.
The flow of most components can be seen in the Component Use diagram [2]. If you are using Piper files or xml Analysis Engine descriptors, to see which components you are using, look at the contents of the Piper files or look at the <fixedFlow> sections of the xml descriptor. [1] https://cwiki.apache.org/confluence/display/CTAKES/cTAKES+4.0+-+Fast+Dictionary+Lookup [2] https://cwiki.apache.org/confluence/display/CTAKES/cTAKES+4.0+Component+Use+Guide#cTAKES4.0ComponentUseGuide-ComponentDependencies On Wed, Dec 20, 2017 at 5:11 AM, satyananda sahu <satyananda...@gmail.com> wrote: > Hi Team, > > > I need some clarification on the process flow of ctakes. Below are my > understanding let me know if I am correct. > > 1. cTAKES uses the UMLS concept as it is for annotating the words from > clinical text. > 2. Based on TUI semantic group, cTAKES has also created semantic group to > hold the set of annotate words. > 3. Each CUI is associated with a TUI, CUI are also linked to snomed and > rxnorm code. CUI is an UMLS concept id. > 3. cTAKES semantic analysis is based on TUI group. > > > Now when I am processing a clinical text, it annotates the word as well as > it gives me all the snomed code associated to the concept. > > someone from team could please help me to identify the place where all this > are happenings > > > Thanks, > Satya >