Standards

 

Back to Bridge2AI Main Page

 

This is a working resource page for the Standards activities of the Bridge2AI program.

The Bridge2AI program will use this wiki to bring together resources relevant to the Standards Modules within the Data Generation Projects and the Standards Core within the BRIDGE Center.

 

Standards Topics

     AI/ML-Readiness 

     Semantic Interoperability

     Inclusion and Data Sufficiency

     Data Bias in AI/ML

     Model Cards

 

Standards Q&A

 

Standards Resources

  • The NESTcc Data Quality Framework focuses primarily on the use of EHR data in the clinical care setting, and considers topics including data governance, characteristics, capture, transformation, and curation.

 

The UMLS Metathesaurus is a large biomedical thesaurus that is organized by concept or meaning. It links synonymous names from over 200 different source vocabularies. The Metathesaurus also identifies useful relationships between concepts and preserves the meanings, concept names, and relationships from each vocabulary. 

The Standards Module experts are encouraged to consult the UMLS Metathesaurus for its adequacy as a standard terminology basis. If it is determined to be inadequate, the applicants should demonstrate the inadequacy and propose an alternative structure that can be aligned with the UMLS Metathesaurus.

The UMLS Metathesaurus can be accessed via the web-based UMLS Metathesaurus Browser (Beta release). Metathesaurus data can also be installed locally using MetamorphoSys, a free tool distributed with the UMLS. A free UMLS Terminology Services account is needed for the access.

 

  • February 2021, ANSI/CTA-2090 The Use of Artificial Intelligence in Health Care: Trustworthiness
  • Standards Warehouse

 

  • ONC’s ISA (Interoperability Standards Advisory) added four social interoperability needs addressing the social determinants of health and a sub section called “care coordination for referrals” and clinical notes
     
  • Logical Observation Identifiers Names and Codes (LOINC) common terminology for laboratory and clinical observations and enables the exchange and aggregation of clinical results for care delivery, outcomes management, and research by providing a set of universal codes and structured names to unambiguously identify things you can measure or observe.
     
  • OMOP Observational Medical Outcomes Partnership used in clinical research - The OMOP Common Data Model allows for the systematic analysis of disparate observational databases. The concept behind this approach is to transform data contained within those databases into a common format (data model) as well as a common representation (terminologies, vocabularies, coding schemes), and then perform systematic analyses using a library of standard analytic routines that have been written based on the common format.
  • SNOMED CT- Systematized Nomenclature of Medicine Clinical Term and is a “common clinical language” consisting of sets of clinical phrases or terms.  It provides a standardized way to represent clinical phrases captured by the clinician and enables automatic interpretation of these and is considered to be the most comprehensive, multilingual clinical healthcare terminology in the world.
     
  • FHIR (Fast Healthcare Interoperability Resources) principle purpose is to deliver data in a common and standard way among independent systems.  It includes data structures, processes, security in one uniform fashion.
     
  • Unified Medical Language System (UMLS) A set of files and software that brings together many health and biomedical vocabularies and standards to enable interoperability between computer systems.
     
  • Value Set Authority Center (VSAC) A repository and authoring tool for standard lists of codes and terms from biomedical vocabularies.
     
  • RxNorm provides normalized names for clinical drugs and links its names to many of the drug vocabularies commonly used in pharmacy management and drug interaction software.
     
  • The lexical tools are a set of computer programs designed to aid in natural language processing. Use Normalized strings.  The three primary programs are:
  • ONC’s ISA (Interoperability Standards Advisory) added four social interoperability needs addressing the social determinants of health and a sub section called “care coordination for referrals” and clinical notes

     

  • Logical Observation Identifiers Names and Codes (LOINC) common terminology for laboratory and clinical observations and enables the exchange and aggregation of clinical results for care delivery, outcomes management, and research by providing a set of universal codes and structured names to unambiguously identify things you can measure or observe.

     

  • Unified Medical Language System (UMLS) A set of files and software that brings together many health and biomedical vocabularies and standards to enable interoperability between computer systems. 

  • Value Set Authority Center (VSAC) A repository and authoring tool for standard lists of codes and terms from biomedical vocabularies.

     

  • RxNorm provides normalized names for clinical drugs and links its names to many of the drug vocabularies commonly used in pharmacy management and drug interaction software.

     

  • The lexical tools are a set of computer programs designed to aid in natural language processing. Use Normalized strings.  The three primary programs are:

  • Standards for AI/ML models :Standards for AL/ML models involve the establishment of conventions for model input format, output format, documentation, and assessment. More information can be found: https://www.iso.org/committee/6794475.html 

Critical Needs

  1. A single, standardized, and comprehensive set of meta-data describing a human patient. This meta-data specification does not cover technology (e.g., imaging, genomic sequencing, ...)
  2. Individual meta-data standards for each technology domain such as imaging, gene expression, etc.
  3. Provenance meta-data to to facilitate credibility of data as it moves through various layers of ownership (data creator, date cleaning, data deposition and archiving, data processing, ...).
  4. Tools to facilitate the creation and validation of the meta-data records defined in the above.

 

 

Table sorting checkbox
On