Synthetic healthcare data utility with biometric pattern recognition using adversarial networks
Abstract This research examines the significance of privacy of synthetic data in healthcare and biomedicine by an analysis of actual data. The significance of authentic health care data necessitates the secure transmission of such data exclusively to authorized users. Therefore, to minimise the reli...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-03-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-94572-3 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract This research examines the significance of privacy of synthetic data in healthcare and biomedicine by an analysis of actual data. The significance of authentic health care data necessitates the secure transmission of such data exclusively to authorized users. Therefore, to minimise the reliance on actual data, synthetic data is developed by incorporating diverse biometric pattern representations, necessitating a distinct setup with adversarial scenarios. Furthermore, to improve the quality of synthetic data, a deep convolutional adversarial network is examined under several operational modes. Furthermore, a distinct conditional metric is employed in this instance to avert the loss of synthetic data, so ensuring consistent transmissions. The system model is developed by examining numerous parameters associated with matching, classification losses, biometric privacy, information leakage, data relocations, and deformations, which are merged with a corresponding adversarial framework. To validate the results of the integrated system model, four scenarios and two case studies are examined, demonstrating that successful data creation can be achieved artificially with minimal losses of 5%. |
|---|---|
| ISSN: | 2045-2322 |