Skip to content

ymnliu/MedTimer

Repository files navigation

Introduction

MedTime is an Apache UIMA based temporal information extraction system developed by the Mayo Clinic NLP program. It extracts and normalizes TIMEX3-based temporal expressions from clinical text. MedTime has been adapted from the open-source temporal tagger, HeidelTime (https://github.com/HeidelTime/heideltime) but re-engineered toward the clinical domain.

What's new

Updated temporal expression extraction module for general purpose clinical texts, which does not require section times or specific text metadata.

Installation

Java requirement: Java 1.7.

If you are using UNIX, simply:

./runCPE.sh

In the menu of the CPE window, go to "File" -> "Open CPE Descriptor" and select ./desc/medtimedesc/collection_process_engine/MedTaggerTimerCPE.xml. You may change the "Input Directory" and "Output Dir" as needed. We suggest to use the default FileCollectionReader and MedTaggerTimerWriter.

Implementation details can be found in project wiki.

Versions

MedTime 1.0: http://ohnlp.org/index.php/MedTime_Project_Page

Citation

If you use MedTime, please kindly consider citing the following papers:

MedTimer 2.0

MedTime 1.0

Other resources

A review paper of clinical information extraction in NLP:

About

Temporal information extraction from clinical texts

Topics

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages