A hierarchical strategy to minimize privacy risk when linking "De-identified" data in biomedical research consortia.
Lucila Ohno-MachadoXiaoqian JiangTsung-Ting KuoShiqiang TaoLuyao ChenPritham RamGuo-Qiang ZhangHua XuPublished in: J. Biomed. Informatics (2023)
Keyphrases
- data sets
- training data
- application domains
- high quality
- data structure
- data analysis
- knowledge discovery
- data processing
- missing data
- raw data
- original data
- prior knowledge
- high dimensional data
- statistical analysis
- image data
- information services
- experimental data
- data points
- databases
- historical data
- data objects
- noisy data
- data acquisition
- synthetic data
- small number
- data sources
- information systems
- machine learning
- data mining