A multiparty homomorphic encryption approach to confidential federated Kaplan–Meier survival analysis
Narasimha Raghavan Veeraragavan, Svetlana Boudko, Jan Franz Nygård · 2026
Read A multiparty homomorphic encryption approach to confidential federated Kaplan–Meier survival analysis by Narasimha Raghavan Veeraragavan, Svetlana Boudko, Jan Franz Nygård (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13645743.txt.
Description
Hospitals can jointly study survival without pooling patient records, Veeraragavan and two colleagues show. Survival analysis across hospitals usually stalls because privacy rules forbid pooling sensitive health data. Simpler sharing methods can leak single counts by subtraction. The team built a federated Kaplan-Meier framework using threshold homomorphic encryption. This method lets a coordinator add coded tallies from many sites. Only the combined result is decoded, using shares from a group.
In tests with synthetic breast-cancer data spread across five hundred sites, the coded curves matched the pooled gold standard closely. Message sizes grew in a steady way with site and time-point counts. The authors also prove correctness, stability, and efficient packing of the encrypted vectors. Veeraragavan and colleagues present the design as a way to run high-quality multi-institution survival studies with far less privacy risk than plaintext collaboration.
Catalog record
| CREATOR | Narasimha Raghavan Veeraragavan, Svetlana Boudko, Jan Franz Nygård |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 329 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |