Perform research and share your results that satisfy GDPR guidelines
by using data anonymization algorithms.
Guarantees exceptional results in the field of Privacy Preserving Data Publishing.
Use Amnesia to transform personal data to anonymous data that can be used for statistical analysis. Data anonymized with Amnesia are *statistically guaranteed* that they cannot be linked to the original data.
Create anonymous datasets from personal data that are treated as statistics by GDPR. Anonymous data can be used without the need for consent or other GDPR restrictions, greatly reducing the effort needed to extract value from them.
Anonymization tailored to user needs through a graphical interface. Guide the algorithm and decide trade-offs with simple visual choices. Developers can incorporate Amnesia anonymization engine to their project through a ReST API.
Get anonymous data in 3 steps
Amnesia accepts complex object relational data in delimited text files.
Visual representations of anonymization parameters and results allow non-expert users to tailor the anonymization process to their needs.
The process is completed without any sensitive data leaving your premises!
Health data for COVID-19 demand anonymization since they contain sensitive personal patients information. Amnesia for Covid-19 is a use case developed for demonstration purposes during the European Commission's EUvsVirus Hackathon.
Amnesia is used in various training courses in data privacy, including the Masters on Business Analytics of AUEB.
Amnesia was used in the EU "My Health My Data" project, to support an ecosystem for safely exchanging medical data. Data included personal details and ICD10 and ICD9 codes for diagnoses. More details here.
To help users quickly familiarize themselves with Amnesia's processes, we released three short tutorial videos to showcase the three main subprocesses of anonymization. The tutorial videos focus on enabling users to understand, tailor and guide the anonymization processes while exploring the quality of the anonymized data.
Better interface for pseudo-anonymization.
Support for user-defined masking rules.
Added error messages on loading dataset
Removed openjfx java library.
Bug fixed on loading Dataverse data.
Privacy guarantees with demographic statistics.
Support for anonymization of DICOM files.