Paper: Temporal Classification of Medical Events

ACL ID W12-2404
Title Temporal Classification of Medical Events
Venue Workshop on Biomedical Natural Language Processing
Session
Year 2012
Authors

We investigate the task of assigning medi- cal events in clinical narratives to discrete time-bins. The time-bins are defined to cap- ture when a medical event occurs relative to the hospital admission date in each clinical narrative. We model the problem as a se- quence tagging task using Conditional Ran- dom Fields. We extract a combination of lexi- cal, section-based and temporal features from medical events in each clinical narrative. The sequence tagging system outperforms a sys- tem that does not utilize any sequence infor- mation modeled using a Maximum Entropy classifier. We present results with both hand- tagged as well as automatically extracted fea- tures. We observe over 8% improvement in overall tagging accuracy with the inclusion of sequence information.