Enhancing the identification of rheumatoid arthritis-associated interstitial lung disease through text mining of chest computerized tomography reports
Authors' conclusion
The addition of ILD-related terms identified by text mining from chest CT reports led to improvements in the PPV of RA-ILD algorithms.
Study context
- Population
- Possible rheumatoid arthritis-interstitial lung disease (RA-ILD) cases from a large academic medical center (derivation) and participants with RA (external validation); medical record review as reference standardderivation 114
- Comparison
- RA-ILD administrative algorithms with vs without ILD-related terms from chest computed tomography (CT) reports
- Outcome
- PPV of RA-ILD algorithms
Methodological detailscross-sectional, derivation and external validation cohorts, natural language processing of chest CT reports
- Cross-sectional
- Derivation and external validation cohorts
- Natural language processing of chest CT reports
Performance
As reported in the abstract
| Test or model | PPV (maximum, derivation cohort) |
|---|---|
| Algorithms with ILD-related terms | 94.6% |
Algorithms with ILD-related terms
PPV (maximum, derivation cohort)94.6%
- PPV improvement with ILD-related terms, derivation cohort
- 3.6 to 11.7%
- PPV improvement with ILD-related terms, validation cohort
- 6.0 to 21.1%
Further findings
Sensitivity, validation cohort
- Decline in sensitivity with ILD-related terms
- −3.9 to −19.5%
Original publication
Enhancing the identification of rheumatoid arthritis-associated interstitial lung disease through text mining of chest computerized tomography reports. Semin Arthritis Rheum. Published 10 April 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →