Identification of Major Bleeding Events in Postoperative Patients With Malignant Tumors in Chinese Electronic Medical Records: Algorithm Development and Validation
Both the LR and CNN methods demonstrate good performance in identifying major bleeding in patients with postoperative malignant tumors from electronic medical records, exhibiting high sensitivity and specificity. Given the higher sensitivity of the LR method (89.47%) and the higher specificity of the CNN method (89.24%) in the test set, both models hold promise for practical application, depending on specific clinical priorities.
Authors
- Changbin Cai (ORCID: https://orcid.org/0000-0003-3706-5185)
- Wei Jiang (ORCID: https://orcid.org/0009-0002-7152-7611)
- Hang Luo (ORCID: https://orcid.org/0009-0004-5154-4718)
- Muhan Yuan (ORCID: https://orcid.org/0000-0001-5398-7396)
- Hui Li (ORCID: https://orcid.org/0000-0003-0579-3858)
- Haiyang Yao
- Yuxiang Gao (ORCID: https://orcid.org/0000-0003-3893-3896)
- Shengming Zhou (ORCID: https://orcid.org/0009-0002-4835-1152)
Institutions
- Beijing Chao-Yang Hospital, Capital Medical University (CN)
- Shanghai Chengtou (China) (CN)
- Sanofi (China) (CN)
- Shanghai Institute of Pharmaceutical Industry (CN)
- Beijing Chaoyang Emergency Medical Center (CN)
Publication Details
- Journal
- JMIR Formative Research
- Published
- 2025-05-01
- DOI
- https://doi.org/10.2196/66189
- Primary Topic
- Machine Learning in Healthcare
- Type
- article
- Field-Weighted Citation Impact
- 0.00