Metadata mapping and reuse in caBIG™
2009

Improving Biomedical Data Interoperability with Mapping Algorithms

Sample size: 66 publication Evidence: moderate

Author Information

Author(s): Isaac Kunz, Ming-Chin Lin, Lewis Frey

Primary Institution: University of Utah

Hypothesis

Can interoperability across biomedical databases be improved by utilizing a repository of Common Data Elements and lexical algorithms?

Conclusion

The study demonstrates that mapping algorithms can effectively reduce the cost and time required to align local data models with reference models in biomedical informatics.

Supporting Evidence

  • The algorithms showed similar performance in mapping UML models to CDEs.
  • The top-ranked matches for both algorithms contained a high percentage of correct mappings.
  • Automated mapping can significantly reduce the manual effort required for data integration.

Takeaway

This study shows that we can use smart computer programs to help connect different databases in medicine, making it easier for researchers to share and use data.

Methodology

The study compared the performance of two algorithms (Dice and Dynamic) in mapping UML model class-attributes to Common Data Elements (CDEs) using similarity measures.

Limitations

The algorithms may struggle with synonyms and abbreviations, which can affect mapping accuracy.

Digital Object Identifier (DOI)

10.1186/1471-2105-10-S2-S4

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication