MT SCENARIO 3

MT SCENARIO 3



MT SCENARIO 3

Background

Text-based patient records are an important resource in medical research. To preserve patient confidentiality, however, the U.S. Health Insurance Portability and Accountability Act (HIPAA) requires that protected health information (GPA) will be removed from the medical records before released. Manual de-identification of large databases of medical records is prohibitive, time consuming and prone to error, necessitating automatic methods for large-scale, automated de-identification.

Technique

We describe an automated Perl-based de-identification software package that is commonly available in most free-text medical records, for example, nursing notes, discharge summaries, reports of X-rays, etc. The software uses lexical look ...