The de-identification of narrative text documents is often realized manually, and requires significant resources.
Here we present a review of recent research in automatic de-identification of narrative text documents in the EHR.
Well aware of these issues, several authors have investigated automated de-identification of narrative text documents from the electronic health record, and a review of recent research in this domain is presented here.
We have reviewed recent research in automatic de-identification of narrative text documents in the EHR, and analyzed the methods and resources used, as well as discussed 18 systems included in this analysis.
Already well aware of these issues, several authors have investigated automated de-identification of narrative text documents from the EHR, and many are described in details and analyzed below.
Although a number of software programs have been developed to address the issue of de-identification of narrative free-text for different types of medical data, [ 6- 17] none have been customized for the full range of primary care EMR notes.
The majority focused only on structured data de-identification instead of narrative text, on image de-identification, or described manual de-identification, and were therefore excluded.
The analysis requires, in the first instance, the identification of local narratives of what threatens and/or undermines the pursuit of a good life.
Through the identification of motifs [13], narrative theories [14] allow us to analyze, interpret, describe, and compare stories across cultures and regions of the world [15].
Charmaz specifically proposed a set of systematic actions that including selecting key words and main native concepts for the identification of meaningful social narratives from thorough analyses of speeches [ 15].
When considering identity construction (i.e. identification) as a narrative process, the role of discursive practices such as metaphor in the construction of new identities (i.e. organizational change) becomes prominent.
