Performance of Natural Language Processing versus International Classification of Diseases Codes in Building Registries for Patients With Fall Injury: Retrospective Analysis
Our findings showed promising performance with higher accuracy of NLP algorithms compared to the conventional method for detecting fall occurrence and mechanism in developing disease registries using clinical notes. Our approach can be introduced to other registries that are based on large data and are in need of accurate annotation and classification.
Authors
- Soheil Ashkani‐Esfahani (ORCID: https://orcid.org/0000-0003-2299-6278)
- Kayhan Batmanghelich (ORCID: https://orcid.org/0000-0001-9893-9136)
- Atta Taseh (ORCID: https://orcid.org/0000-0001-7485-1500)
- Ara Nazarian (ORCID: https://orcid.org/0000-0003-3992-3489)
- Jonathan F. Bean (ORCID: https://orcid.org/0000-0001-8385-8210)
- M. Chan
- Evan Sirls (ORCID: https://orcid.org/0009-0002-5037-0061)
- Souri Sasanfar
Institutions
- Boston University (US)
- Beth Israel Deaconess Medical Center (US)
- Harvard University (US)
- Spaulding Rehabilitation Hospital (US)
- Massachusetts General Hospital (US)
- Geriatric Research Education and Clinical Center (US)
Publication Details
- Journal
- JMIR Medical Informatics
- Published
- 2025-02-21
- DOI
- https://doi.org/10.2196/66973
- Primary Topic
- Machine Learning in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 0.00