Anonymization of Clinical Data Using Time-Sensitive Methods
Author Information
Author(s): Hammer Freimut, Buglowski Mateusz, Stollenwerk André
Primary Institution: RWTH Aachen University
Hypothesis
Can a method for anonymizing time-continuous clinical data be developed that preserves the relationship between time and value while ensuring privacy?
Conclusion
The proposed method effectively anonymizes clinical data while minimizing information loss and maintaining privacy.
Supporting Evidence
- The method preserves the dynamics of time-continuous data.
- It reduces information loss while ensuring privacy.
- The approach is based on established privacy measures like k-anonymity and t-closeness.
- Data was evaluated from animal experiments to ensure practical applicability.
Takeaway
This study created a way to hide personal information in medical data while still keeping the data useful for research.
Methodology
The method involves splitting data along the time-axis, using bucketization and distribution clustering to achieve k-anonymity and t-closeness.
Potential Biases
Potential risks include the possibility of re-identification if the data is not properly anonymized.
Limitations
The method may not be suitable for all types of data and requires careful parameter selection.
Participant Demographics
Data was collected from animal trials, specifically focusing on respiratory data.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website