This project aims to create integrated models that can effectively analyze a wide range of clinical notes, such as consultations, discharge summaries, echocardiography reports, history and physical documents, operative reports, progress notes, selection conference records, and social worker notes to predict whether a patient is likely to be re-hospitalized for kidney transplant.
- PI: Rachel Patzer - Associate Professor of Surgery, Emory University
- NLP: Jinho Choi - Assistant Professor of Computer Science, Emory University
- Predicting Kidney Transplant Recipient Cohorts’ 30-Day Re- Hospitalization Using Clinical Notes and Electronic Healthcare Record Data. Arenson, M.; Hogan, J.; Xu, L.; Lynch, R.; Lee, Y. H.; Choi, J. D.; Sun, J.; Adams, A.; Patzer, R. Kidney International Reports (KIR), 2023.
- Noise Pollution in Hospital Readmission Prediction: Long Document Classification with Reinforcement Learning. Xu, L.; Hogan, J.; Patzer, R. E.; and Choi, J. D. Proceedings of the ACL Workshop on Biomedical Natural Language Processing (BioNLP), 2020.
- Multimodal Ensemble Approach to Incorporate Various Types of Clinical Notes for Predicting Readmission. Shin, B.; Hogan, J.; Adams, A. B.; Lynch, R. J.; Patzer, R. E.; and Choi, J. D. Proceedings of the IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI), 2019.