Extracting LOINC Codes from a Laboratory Information System's Index: Addressing Semantic Interoperability with Web...

Area of Science:

• Medical Informatics
• Bioinformatics
• Health Data Standards

Background:

• Laboratory information systems (LIS) often use internal codes, hindering data interoperability and secondary use in research and healthcare.
• Standardizing laboratory data is crucial for seamless data exchange and integration into health data infrastructures.
• Lack of standardized terminologies in LIS impedes queryability and data aggregation.

Purpose of the Study:

• To automate the extraction and mapping of internal laboratory codes to LOINC (Logical Observation Identifiers Names and Codes).
• To enhance structured data integration by creating a FHIR-compliant ConceptMap.
• To address the challenge of non-standardized laboratory data in LIS.

Main Methods:

• Developed a Python-based workflow utilizing Selenium, BeautifulSoup, and Pandas for data extraction and processing.
• Extracted laboratory data from an internal lab index.
• Mapped extracted internal codes to LOINC and generated a FHIR-compliant ConceptMap.

Main Results:

• Successfully extracted 2,870 analytes from the internal lab index.
• Mapped 768 analytes (27%) to LOINC codes.
• Demonstrated the feasibility and scalability of the automated mapping process.

Conclusions:

• The automated approach facilitates structured laboratory data integration.
• Highlights the necessity for direct integration with LIS for comprehensive data standardization.
• Emphasizes the need for expanded LOINC coverage, particularly for legacy data, to maximize interoperability.