Continuity of Care Documents (CCD) Case Study
Continuity of Care Documents (CCD) Case Study
This section presents a case study where we extract data from CCD (Continuity of Care Documents).
This case study illustrates that how the NLP API can leverage known document standards such the Clinical Document Architecture (CDA), a popular and flexible document standard, to retrieve and organize data for many clinical and research purposes. In this case study, a custom document type plugin, and associated engine extensions were implemented to extract and store entities extracted from unstructured/semi-structured text elements as well as structured data entries found in CCD document sections. Because some system-specific assumptions were made, the approach taken to extract and query the structured output may not directly apply to your use case.
If you have a use case that involves processing of CCD documents, or other types of CDA-based document templates, please contact us at support@emtelligent.com.
About the sample documents. The queries in these sections were tested against from four publicly available C-CDA sample documents, processed into a database of synthetic clinical records. They come from two well-known open collections of C-CDA samples:
- jmandel/sample_ccdas —
Vitera/Vitera_CCDA_SMART_Sample.xmlandEMERGE/Patient-1.xml - jddamore/ccda-samples —
Get Real Health/CCD_e1.xmland360 Oncology/Alice_Newman_health_summary Delegate.xml
These were chosen because between them they carry populated Results, Vital Signs, Immunizations and Social History sections, which the queries below rely on. They contain no real patient data, and the patient names in the output (Alice Newman and others) are the standard C-CDA test personas rather than anyone’s records.

