UMLS-Core: SMT Requirements
- Objective:
- Develop a set of software that map a term (multiple words) to CUI
- Software Specification:
- A command-line software to provides functions described in the next section
- Functions:
- Inputs:
- list of terms:
- clinterms0610.xlsx (only use 1st field)
- VATerm.csv (use 1st & 2nd field)
- KP_SynMap_out.txt
- remove header (1 row)
- Synonyms Standardization
- Term to CUI mapping:
- MRXNS_ENG.RRF & MRCONSO.RRF
- only use 2nd and 3rd fields
- remove duplications
- exclude suppressible atoms
- Preferred term:
- Outputs:
- Format:
Term ID | Original Term | Matching term (normalized with synonyms) | CUI | Flag | Preferred Terms
|
where:
- Flag:
Value | Description
|
---|
0 | CUI found, with no replaced synonym
|
1 | CUI found, with one replaced synonym
|
2 | CUI found, with two replaced synonyms
|
3 | no CUI found (C0000000)
|
- Processes:
- Pre-Processes:
- Non-Suppressible atoms:
- MRCONSO.RRF
- SUPPRESS=N (non-suppressible)
- SUPPRESS=O (suppressible)
- SUPPRESS=E (suppressible)
- SUPPRESS=Y (suppressible)
- Outputs: CUI|LUI|SUI
- Preferred Terms:
- MRCONSO.RRF
- LAT=ENG
- TS=P
- STT=PF
- ISPREF=Y
- Only one preferred term per CUI
- Outputs: CUI|preferred term
- Processes:
- normalize input term (use norm) and map to CUI (from MRXNS_ENG.RRF)
- if no CUI found from above, replace input term with one synonym and map to CUI
- if no CUI found from above, replace input term with two synonyms and map to CUI
- if no CUI found from above, use "orginal term|original term|C000000|3
- print out results
- Post-Processes:
- Add preferred term to the result file (by CUI|preferred term mapping)
- Examples:
- KP103521|HIV positive|hiv positive|C0019699|0|HIV Seropositivity
- KP103221|EXAM, FOOT, DIABETIC|diabetic examination foot|C1303165|1|Diabetic foot examination
- KP168981|HX OF VARICELLA|chickenpox history|C0455469|2|H/O: chickenpox
- VA1000|Alcohol Withdrawal|alcohol withdrawal|C0236663|0|Alcohol withdrawal syndrome
- VA10137|Leg Condition|infection leg|C0745687|1|leg; infection
- VA14760|Pneumonia due to other virus|other pneumonia viral|C0348677|2|Other viral pneumonia