Skip to main content
Fig. 1 | BMC Medical Informatics and Decision Making

Fig. 1

From: Accurate training of the Cox proportional hazards model on vertically-partitioned data while preserving privacy

Fig. 1

Performance of the matrix inverse protocol. This figures demonstrates the scalability of the matrix inverse in the number of covariates (dimension of the matrix). The filled data points are based on an average of 100 runs per datapoint. The open data points are based on a single run. We need to perform one matrix inversion per iteration of the secure CPH protocol. Remark: the current implementation supports matrix inversions of matrix sizes of \(2\times 2\) upto \(14\times 14\)

Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us