Publications by authors named "Jinkyu Yoo"
JMIR Med Inform
November 2021
Article Synopsis
- The study focuses on developing improved event prediction models for intensive care units, addressing issues like temporal skewness in data from electronic medical records and the potential for errors and delays in input.
- Researchers analyzed data from 21,738 patients to predict three critical events: death, sepsis, and acute kidney injury, using multiple models designed to enhance prediction accuracy and robustness against errors.
- Results indicated that the new models generally outperformed traditional methods, showing better accuracy, especially under conditions of simulated input errors, while maintaining performance even with delayed information.
