Shared resources

field description of body part mapping file:

1. term id - T are title terms from the definition table; I are inclusion terms from definition table, X are terms used in the PCS tables but not included in definition file
2. body part term
3. map method - either using MetaMap or normalized string match
4. SNOMED CT concept id
5. SNOMED CT fully-specified name
    Statistics: total PCS terms 1917 of which 1580 have SNOMED CT map (1383 by MetaMap, 1305 by normalized string match) 

The two files SNOMED Bodysite values distribution.txt and SNOMED Method values distribution.txt are counts of concepts that used these values in the Procedure site - direct attribute and Method attribute. There are two counts, one for all procedures and one for only surgical procedures.

PCS_BODY_PART_DEFINITION_TABLE.xlsx file has all the body part names and body part keys in the PCS body part definition table. Note that it's 2016 (not 2017) version of PCS.

PCS_AXIS4_NOT_IN_DEF_TABLE.xlsx file has the body part values that are used in the axis4 of PCS codes but not included in the body part definition table. To get all body part values, we need to combine this table with the body_part_definition table.

The Axis_3_And_4_Map Access file goes with the ppt presentation of Nov 17. Each object has some comment to it when you right click and open the 'Table properties' 

Mapeo-eje-4-cie10-pcs-a-snomed-ct-20161219.xlsx file is the new version of axis #4 reverse mapping. It's include genitourinary, gastrointestinal, hepatobiliary, and mouth and throat systems

mapeo-eje-4-cie10-pcs-a-snomed-ct-20161219-consolidado.xlsm file is a consolidated version of all PCS Axis 4 reverse mappings available up to the specified date 20161219; as new body systems are added to the reverse map, we will consolidate map elements into one table; some metrics will be added and updated to monitor progress. Thanks to the team working from Cordoba, Argentina.

Mapeo-eje-4-cie10-pcs-a-snomed-ct-20170111.xlsx file is the new version of axis #4 reverse mapping. It's adds endocrine system.

Reverse mapping for axis 4 mapeo-eje-4-cie10-pcs-a-snomed-ct-20170112-consolidado.xlsm has been updated. Endocrine system appended. 

Ontological mapping.xlsm contains a reverse mapping table for axis 3 of PCS. It requires additional normalization for use within queries. Ready for discussion.

PCS_Ontology_2016.xlsm contains a bilingual version of PCS, English and Spanish.

I've uploaded a new version of the ICD9CM procedure index mapping file (ICD9CM index map to SNOMED and PCS 20170925.zip). I've added a new field called CONCATNAME_SHORT (short concatenated name), which is the concatname without the words in parenthesis (non-essential modifiers) and other things not useful in mapping, such as NEC, NOS, and other * (e.g. other specified site). This results in many more hits (50% increase) when mapping to SNOMED CT using normalized lexical mapping. The field list:

ID - master ID of the index entry

Main - main entry of index

indent1-indent7 - indented entries

code1 - first code listed

code2 - second code listed

Ref - reference

concatname - concatenation of main entry and indented entries

CONCATNAME_SHORT - concatenation of main entry and indented entries, ignoring words in parenthesis, NEC, NOS, and other * (e.g. other specified site)

NORMMAP_CONCEPTID - map to SNOMED CT conceptid using normalized lexical matching of concatname

NORMMAP_FSN - fully specified name of the SNOMED CT concept

NORMMAP_SHORT_CONCEPTID - map to SNOMED CT conceptid using normalized lexical matching of CONCATNAME_SHORT

NORMMAP_SHORT_FSN - fully specified name of the NORMMAP_SHORT_CONCEPTID

UMAP_CODE1_CONCEPTID - map to SNOMED CT conceptid using UMLS synonymy and first ICD9CM code (code1)

UMAP_CODE1_FSN - fully specified name

UMAP_CODE2_CONCEPTID - map to SNOMED CT conceptid using UMLS synonymy and second ICD9CM code (code2)

UMAP_CODE2_FSN - fully specified name

CODE1_GEM_PCS - map to PCS using GEM based on first ICD9CM code (code1)

CODE1_GEM_FLAG - flag value of first GEM map

CODE2_GEM_PCS - map to PCS using GEM based on second ICD9CM code (code2)

CODE2_GEM_FLAG - flag value of second GEM map

A new version of MAGPIE is ready to try (magpie-0.2.0-full.jar).

New instructions:

Running the MAGPIE server - from the file explorer

1. Your PC will need Java 8.

2. Make a folder in which to put MAGPIE and its data.  Allow at least

    500 megabytes.

3. Put the MAGPIE jar file in that folder.

4. Double-click the MAGPIE jar file (or otherwise invoke it, according

    to the customs of your computer desktop).

    - Soon, a picture of the Lister Hill National Center for Biomedical

      Communications should appear.  Java is initializing.

    - After a little while, the MAGPIE Control box appears, with a

      message "Preparing...".  MAGPIE is reading its data files and

      expanding them into a searchable index.

    - After a few minutes, the MAGPIE Control box says, "Serving

      http://localhost:8080/".  There are two buttons, "Browse" and

      "Stop".  MAGPIE is now ready to serve its website.

8. Use the MAGPIE website: Choose "Browse", or by other means open a

    web browser to the location http://localhost:8080/

9. Stop the MAGPIE server when you are finished.

    You may do this by choosing Stop from the MAGPIE Control box.

10. To run MAGPIE again, repeat the process from Step 4,

     "double-click the MAGPIE jar file...".  MAGPIE will re-use the

     expanded index of data files, saving some time relative to the

     initial run. 

This file OMOPUV5_201901_PCS_SCT_MAP.csv contains the maps between SNOMED CT and PCS extracted from the OHDSI's terminology tables. This may be useful for batch matching between the two terminologies. The fields are:

conceptid1 - internal OHDSI concept id for the PCS code

PCSname

PCScode

relationship_id - all the rows are Is_a relationships

conceptid2 - internal OHDSI concept id for the SNOMED CT concept

SCTname

SCTcode

This file OMOPUV5_201901_PCS_SCT_MAP_ALL.csv contains all relationships, not only Is_a.

Arturo,

I don't think it is a problem to use it, but to be safe do contact them https://ohdsi.org/.

Hi there,

I've uploaded a zip file called ICHI. It contains content scraped from the ICHI browser, since no flat file is available to download. There are 2 sets of files:

1. extension.xx - these cover the extension codes. The extension_withTop.xlsx file has all the extension codes and the attributes that only have one value (e.g. title, definition), the other files are those attributes that can have more than one value (you need to join them to the extension file by the ExtensionCode). The extension_hierarchy_withcounts file shows all extension codes with their top level ancestors and descendant counts.
2. ichi.xx - cover all intervention codes. The intervensions_withTop.xlsx file has all the intervention codes and the attributes that only have one value (e.g. ICHI descriptor, definition), the other files are those attributes that can have more than one value (you need to join them to the extension file by the ICHI Code).  the file ichi_interventions_hierarchy_withcounts shows all extension codes with their top level ancestors and descendant counts
   
   

Let me know if you have questions.

New ICHI (Beta 3 version) flat files for interventions and extension codes uploaded