Abstract
Acute kidney injury (AKI) in critically ill patients is associated with significant morbidity and mortality. Development of novel methods to identify patients with AKI earlier will allow for testing of novel strategies to prevent or reduce the complications of AKI. We developed data-driven prediction models to estimate the risk of new AKI onset. We generated models from clinical notes within the first 24 hours following intensive care unit (ICU) admission extracted from Medical Information Mart for Intensive Care III (MIMIC-III). From the clinical notes, we generated clinically meaningful word and concept representations and embeddings, respectively. Five supervised learning classifiers and knowledge-guided deep learning architecture were used to construct prediction models. The best configuration yielded a competitive AUC of 0.779. Our work suggests that natural language processing of clinical notes can be applied to assist clinicians in identifying the risk of incident AKI onset in critically ill patients upon admission to the ICU.
| Original language | English (US) |
|---|---|
| Title of host publication | Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018 |
| Editors | Harald Schmidt, David Griol, Haiying Wang, Jan Baumbach, Huiru Zheng, Zoraida Callejas, Xiaohua Hu, Julie Dickerson, Le Zhang |
| Publisher | Institute of Electrical and Electronics Engineers Inc. |
| Pages | 683-686 |
| Number of pages | 4 |
| ISBN (Electronic) | 9781538654880 |
| DOIs | |
| State | Published - Jan 21 2019 |
| Event | 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018 - Madrid, Spain Duration: Dec 3 2018 → Dec 6 2018 |
Publication series
| Name | Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018 |
|---|
Conference
| Conference | 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018 |
|---|---|
| Country | Spain |
| City | Madrid |
| Period | 12/3/18 → 12/6/18 |
Fingerprint
Keywords
- Machine Learning
- Medical Decision Making
- Natural Language Processing
- Unified Medical Language System
ASJC Scopus subject areas
- Biomedical Engineering
- Health Informatics
Cite this
}
Early Prediction of Acute Kidney Injury in Critical Care Setting Using Clinical Notes. / Li, Yikuan; Yao, Liang; Mao, Chengsheng; Srivastava, Anand; Jiang, Xiaoqian; Luo, Yuan.
Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018. ed. / Harald Schmidt; David Griol; Haiying Wang; Jan Baumbach; Huiru Zheng; Zoraida Callejas; Xiaohua Hu; Julie Dickerson; Le Zhang. Institute of Electrical and Electronics Engineers Inc., 2019. p. 683-686 8621574 (Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
TY - GEN
T1 - Early Prediction of Acute Kidney Injury in Critical Care Setting Using Clinical Notes
AU - Li, Yikuan
AU - Yao, Liang
AU - Mao, Chengsheng
AU - Srivastava, Anand
AU - Jiang, Xiaoqian
AU - Luo, Yuan
PY - 2019/1/21
Y1 - 2019/1/21
N2 - Acute kidney injury (AKI) in critically ill patients is associated with significant morbidity and mortality. Development of novel methods to identify patients with AKI earlier will allow for testing of novel strategies to prevent or reduce the complications of AKI. We developed data-driven prediction models to estimate the risk of new AKI onset. We generated models from clinical notes within the first 24 hours following intensive care unit (ICU) admission extracted from Medical Information Mart for Intensive Care III (MIMIC-III). From the clinical notes, we generated clinically meaningful word and concept representations and embeddings, respectively. Five supervised learning classifiers and knowledge-guided deep learning architecture were used to construct prediction models. The best configuration yielded a competitive AUC of 0.779. Our work suggests that natural language processing of clinical notes can be applied to assist clinicians in identifying the risk of incident AKI onset in critically ill patients upon admission to the ICU.
AB - Acute kidney injury (AKI) in critically ill patients is associated with significant morbidity and mortality. Development of novel methods to identify patients with AKI earlier will allow for testing of novel strategies to prevent or reduce the complications of AKI. We developed data-driven prediction models to estimate the risk of new AKI onset. We generated models from clinical notes within the first 24 hours following intensive care unit (ICU) admission extracted from Medical Information Mart for Intensive Care III (MIMIC-III). From the clinical notes, we generated clinically meaningful word and concept representations and embeddings, respectively. Five supervised learning classifiers and knowledge-guided deep learning architecture were used to construct prediction models. The best configuration yielded a competitive AUC of 0.779. Our work suggests that natural language processing of clinical notes can be applied to assist clinicians in identifying the risk of incident AKI onset in critically ill patients upon admission to the ICU.
KW - Machine Learning
KW - Medical Decision Making
KW - Natural Language Processing
KW - Unified Medical Language System
UR - http://www.scopus.com/inward/record.url?scp=85060857247&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85060857247&partnerID=8YFLogxK
U2 - 10.1109/BIBM.2018.8621574
DO - 10.1109/BIBM.2018.8621574
M3 - Conference contribution
T3 - Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018
SP - 683
EP - 686
BT - Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018
A2 - Schmidt, Harald
A2 - Griol, David
A2 - Wang, Haiying
A2 - Baumbach, Jan
A2 - Zheng, Huiru
A2 - Callejas, Zoraida
A2 - Hu, Xiaohua
A2 - Dickerson, Julie
A2 - Zhang, Le
PB - Institute of Electrical and Electronics Engineers Inc.
ER -