disi-unibo-nlp/physionet-deid-i2b2-2014
The De-identification dataset contains medical text records with Named Entity Recognition (NER) annotations. The dataset is processed to split records into individual sentences while preserving entity annotations. Each sentence is tokenized and annotated in IOB format for training NER models.
Labels for Physionet Gold Corpus following the I2B2-2014 de-identification challenge guidelines ==========
The data consists of labels for the Physionet Gold Corpus dataset: https://physionet.org/content/deidentifiedmedicaltext/1.0/.
Annotation follows the guidelines described in section 4 of: Amber Stubbs OU. Annotating longitudinal clinical narratives for de-identification: the 2014 i2b2/UTHealth Corpus. J Biomed Inform. 2015;58(Suppl):S20, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4978170/.
This labeling was used in the paper "Customization Scenarios for De-identification of Clinical Notes" in order to generate results for models trained on I2B2-2014 dataset, and evaluated on the Physionet Gold Corpus dataset.
Contact: ------- For questions or comments, please contact corresponding authors appearing in the paper.
Data: ---- Data includes:
- I2B2-2014-Relabeled-PhysionetGoldCorpus.csv Comma Separated Values with the following columns: a. recordid - string of the form "<int>||||<int>||||" corresponding to the id given in the original Physionet Gold Corpus dataset. For example, "22||||35||||" corresponds to STARTOFRECORD=22||||35|||| start marker in the original Physionet corpus. b. begin - int, specifying the relative offset of the span relative to STARTOF_RECORD. c. length - int, specifying the length of span. d. type - enum, with one of the following values: NAME, AGE, ORGANIZATION, ID, DATE, CITY, STATE, COUNTRY, LOCATION, HOSPITAL, PHONE, PROFESSION.