SNOMED CT Templates

Page tree

StatusIn PROD
Version

2.0

Descriptions:

Termdescription typeLanguage/acceptabilityLanguage/acceptabilityCase significance

[course] [periods of life] [morphology] of [body structure] caused by [bacteria] (disorder)

FSNus:Pgb:Pci

[course] [periods of life] [morphology] of [body structure] caused by [bacteria]

SYNus:Pgb:Pci


Concept model:

Attribute CardinalityAttributeValuerole group numberrole group cardinality
1..1

116680003 |Is a (attribute)|

64572001 |Disease (disorder)|

0N/A
0..1

263502005 |Clinical course (attribute)|  

[course] <  288524001 |Courses (qualifier value)|

N/A

1..1

370135005 |Pathological process (attribute)|

441862004 |Infectious process (qualifier value)|

1

1..1



0..1

363698007 |Finding site (attribute)|

[body structure] <<  442083009 |Anatomical or acquired body structure (body structure)|

0..1

116676008 |Associated morphology (attribute)|

[morphology] <<  49755003 |Morphologically abnormal structure (morphologic abnormality)|

1..1

246075003 |Causative agent (attribute)|

[Bacteria] << 409822003 |Superkingdom Bacteria (organism)|

0..1

246454002 |Occurrence (attribute)|

[periods of life] <  282032007 |Periods of life (qualifier value)|

Definition status:  


900000000000073002 |Defined (core metadata concept)|

Applies to:

<<  87628006 |Bacterial infectious disease (disorder)|  

Rules for description generation:

  1. Apply general rules Template for simple template of disease - v1.0;
  2. Replace [morphology] by the word 'infection' if associated morphology is absent; 
  3. Replace [morphology] by the word 'infection' if associated morphology is 409774005|Inflammatory morphology (morphologic abnormality)| or 23583003|Inflammation (morphologic abnormality)|. 

JIRA ticket:

INFRA-2999 - Getting issue details... STATUS

18 Comments

  1. Jim Case

    Do we really need this one if all we are doing is eliminating Inflammation, but retaining all of the subtypes?  I was more thinking of morphologies that are not transitive to Inflammation

    1. Yongsheng Gao

      Hi Jim Case, this template does have [morphology] where morphologies are not inflammatory. The specific morphology needs to be represented in all descriptions. For example, bacterial cellulitis, the morphology is cellulitis.

      I am still looking into those morphologies that are not transitive to inflammation. Most of them are appeared to be a disorder "with infection". Some of them are modelled by two role groups and some are in a single role group. So, we need to determine the modeling pattern for them. I will check with Bruce Goldberg for the progress on this subject.

      The other template is solely for 'infection' that can be modelled with or without inflammatory morphology OR inflammation. However, the morphology of inflammation is "silent" in the descriptions. We do not include [morphology] in description template. 

      1. Bruce Goldberg

        Yongsheng Gao, Jim Case. I have just begun looking at this. There are a number of modeling patterns. Modeling with 2 role groups may be encountered when the infection occurs at > 1 body structure (e.g. vulvovaginitis). There are a number of osteomyelitis concepts modeled with infectious process, inflammation and finding sites of both bone and bone marrow. I wrote up a proposed solution for remodeling osteomyelitis in the past. Another cause of > 1 role group is that many concepts are modeled with fully defined parents. Another issue is that in 1 instance the disorder was modeled as in infection but is really due to an infection.

        I created the following ECL query (82 hits) which I think captures those concepts with morpholgies not inflammation or a subtype. Again in some cases, the concepts are modeled with parents of the infection and the result of the infection leading the concept to inherit the morphology of the result

        << 40733004 |Infectious disease|:[2..*]{
        [1..1] 370135005 |Pathological process (attribute)|=*}
        AND [0..0] 42752001|Due to (attribute)|=*
        AND [0..0] 255234002|After (attribute)|=*
        AND [0..0] 47429007|Associated with (attribute)|=*
        AND [1..*] 116676008 |Associated morphology (attribute)|!=<<23583003 |Inflammation (morphologic abnormality)|

        P.S. can I receive permission to post files here?

        Bruce

  2. Jim Case

    I may be getting lost in the extensive discussion, but I think it is important to recognize that the concept 23583003 |Inflammation (morphologic abnormality)| is going to go away as it conflates structure and process.  Not sure how that affects the template at this stage, but it is an important consideration.

  3. Yongsheng Gao

    Michael Chu We could merge two template into one if the rules 4 - 6 for descriptions can be implemented. Please let us know if it is possible. 

    1. Michael Chu

      Yongsheng Gao We need to implement new features to support rules 4-6. Yes it is possible.

      Rule 4 above "Caused by [bacteria]" should be removed from descriptions if causative agent is not present in the concept model." doesn't seem to match the [1..1] cardinality of causative agent attribute in the logic model. Can you please check and confirm? Thanks.

      1. Yongsheng Gao

        Hi Michael Chu, I have added some general rules for descriptions related to optional attributes. Please see detail at Template for simple template of disease. So, we might not need to repeat these rules in each individual template. Let's discuss the option this morning. 

        1. Michael Chu

          Yongsheng Gao General rules for removing term when option slot is not present are fine. Please note when the cardinality is 1..1 as shown in the logical model for [Bacteria] the removing rule will not apply. 

          1. Yongsheng Gao

            Michael Chu agreed, removal is only applicable to optional attributes.

  4. Yongsheng Gao

    Hi Jim Case, I have added three new rules for generating descriptions. So, we can have a single template for bacterial infection. If it is okay, I will change the status to ready for implementation.

    1. Jim Case

      Please do so...

  5. Michael Chu

    Yongsheng Gao I feel that we don't need to state rule 4 explicitly as it is implied by default. However rule 3 is required.

    1. Yongsheng Gao

      Cheers Michael Chu  agreed and removed rule 4.  I just added the role group cardinality. 

  6. Peter G. Williams

    Yongsheng Gao  picking up the work to keep our KPI stats up (see https://docs.google.com/spreadsheets/d/1lL9vOTXRQNT_oUxtjkOZEupmikl8GZXL_-2i5ddc2C0/edit#gid=1 ) I'm looking to see if the template I have in my local library is fully covered by what we have in production.   Both sets of templates are checked, but no double counting is allowed - ie if a concept aligns to any template then it is counted, but only once.

    This template has an optional finding site, but the domain that the template applies to is <<  301811001 |Bacterial infection by site (disorder)|  

    Could we in fact apply this template to << 87628006 |Bacterial infectious disease (disorder)|  and apply it to 2405 concepts rather than 1727 as is currently scoped?

    1. Jim Case

      Peter G. Williams,

      I would agree that this template can apply to the concept 87628006 |Bacterial infectious disease (disorder)|.  As a non-clinical grouper, 301811001 |Bacterial infection by site (disorder)| is under evaluation for inactivation anyway as finding by site is easily handled with ECL.

      1. Peter G. Williams

        Super.   Are we happy to just push that change through without having to increment the template version?   I can raise a TBBA ticket if so.


        Update TBBA-450 raised.

        1. Yongsheng Gao

          I agree with Jim. It will be fine if the technical team is okay to make the changes in the authoring platform. So, we do not need to create a new Confluence page. 

          1. Jim Case

            All sounds great to me...