Deep Learning Models to Screen Electronic Health Records for Breast and Colorectal Cancer Progression: Performance Evaluation Study
The deep learning language models could help identify breast and colorectal cancer progression in EHRs and remove most charts from the chart review process. A limited number of tokens may influence model predictions. Improvements in model performance could be obtained by increasing the training dataset size and analyzing EHRs at the sentence level rather than at the EHR level.
Authors
- Marshall Pitz (ORCID: https://orcid.org/0000-0002-1858-5081)
- Helen Chen (ORCID: https://orcid.org/0000-0002-9401-7895)
- Harminder Singh (ORCID: https://orcid.org/0000-0002-9354-2356)
- Pascal Lambert (ORCID: https://orcid.org/0000-0003-2126-7591)
- Rayyan Azam Khan (ORCID: https://orcid.org/0000-0001-5816-1251)
- Kathleen Decker (ORCID: https://orcid.org/0000-0002-0044-7545)
Institutions
- University of Waterloo (CA)
- CancerCare Manitoba (CA)
- Research Institute in Oncology and Hematology (CA)
- University of Manitoba (CA)
Publication Details
- Journal
- JMIR AI
- Published
- 2025-10-13
- DOI
- https://doi.org/10.2196/63767
- Primary Topic
- Machine Learning in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 0.00