JMIR Publications

ABSTRACT

Background:

Valuable insights gathered by clinicians during their inquiries and documented in textual reports are often unavailable in the structured data recorded in the electronic health records (EHRs).

Objective:

This work highlights that mining unstructured textual data with natural language processing (NLP) techniques complements the available structured data and enables a more comprehensive patient phenotyping.

Methods:

We collected EHRs available in the clinical data warehouse of the Greater Paris University Hospitals from 2012 to 2021 for patients hospitalized and diagnosed with one of four immune-mediated inflammatory diseases: systemic lupus erythematosus (SLE), systemic sclerosis, antiphospholipid syndrome, and Takayasu's arteritis. Then, we built, trained, and validated NLP algorithms on 103 discharge summaries selected from the cohort and annotated by a clinician. Finally, all discharge summaries in the cohort were processed with the algorithms and the extracted data were compared with the structured data.

Results:

Named entity recognition followed by normalization yielded f1-scores with 95% confidence intervals of 71.1 (63.6-77.8) for the laboratory tests and 89.3 (85.9-91.6) for the drugs. Application of the algorithms to 18,604 EHRs increased the detection of antibody results and drug treatments (e.g., +53.6% of the patients in the SLE cohort with positive antinuclear antibodies), making the results more consistent with the literature.

Conclusions:

While challenges remain in standardizing laboratory tests, particularly with abbreviations, this work, based on secondary use of clinical data, demonstrates that automatic processing of discharge summaries enriched the information available in structured data and facilitated more comprehensive patient profiling.