Datafly: a system for providing anonymity in medical data

摘要

We present a computer program named Datafly that maintains anonymity in medical data by automatically generalizing,substituting,inserting and removing informaiton as appropriate without losng many of the details found within the data.Decisions are made at the field and record level at the time of database access,so the approach can be used on the fly in role-based security within an institution,and in batch mode for exporting data from an institution.Ofte organizations release and receive medical data with all explicit identifiers,such as name,address,phone number,and Social Security number,removed in the resulting data look anonymous; howwever,we shown that in most of these cases,the remaining data cna be used to re-identify individuals by linking or matching the data to other databases or by looking at unique characteristics found in the rifeld and records of the database itself.When these less apparent aspects are takent ito account,each released record cna be made to ambiguously map to many possble people,providing a levle of anonymity which the user determines.