Sign in Subscribe
Natural language processing

Natural Language Processing System Improves Electronic Medical Records Quality

Strategic Insights

— 1 min read

Current study results on Information Technology - Electronic Medical Records have shown that well-organised electronic health records are essential for high-quality patient care. However, electronic health records user interfaces can be cumbersome for entry of structured information, resulting in a majority of information being in free text rather than a structured form. Researchers from University College London (UCL) have developed an open-source, modular, and configurable natural language processing (NLP) system called MiADE, which integrates with an electronic health record system. The MiADE system includes components to extract diagnoses, medications, and allergies from a clinical note, and communicate with an electronic health record system in real-time using Health Level 7 Clinical Document Architecture (HL7 CDA) messaging.

Key Takeaways:

  • The MiADE system achieved a precision of 83.2% (95% CI 77.0, 88.1) and recall of 85.2% (95% CI 79.1, 89.8) for detection of diagnosis concepts using the MedCAT library for named entity recognition (NER) and linking to SNOMED CT, as well as context detection.
  • The system reduced the time taken for clinicians to enter structured problem lists by 89% in simulation testing.
  • The MiADE system has commenced a trial implementation at University College London Hospitals (UCLH) integrated with the Epic EHR.
  • The system was developed as part of a collaboration between researchers from UCL and the Institute of Health Informatics.
  • The open-source MiADE system is designed to be modifiable and configurable to meet the needs of different healthcare institutions.

Statistics:

  • 83.2% precision (95% CI 77.0, 88.1) for detection of diagnosis concepts
  • 85.2% recall (95% CI 79.1, 89.8) for detection of diagnosis concepts
  • 89% reduction in time taken for clinicians to enter structured problem lists in simulation testing

Sources:

  • "Design and implementation of a natural language processing system at the point of care: MiADE (medical information AI data extractor)" published in BMC Medical Informatics and Decision Making, 2025;25(1):365.
  • "MiADE system integrated with the Epic EHR at University College London Hospitals" reported by University College London (UCL)
  • Institute of Health Informatics, University College London (UCL)
  • Epic Systems Corporation
  • Health Level 7 International (HL7)

Read more