English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Survival analysis is a fundamental tool in medicine, modeling the time untilan event of interest occurs in a population. However, in real-worldapplications, survival data are often incomplete, censored, distributed, andconfidential, especially in healthcare settings where privacy is critical. Thescarcity of data can severely limit the scalability of survival models todistributed applications that rely on large data pools. Federated learning is apromising technique that enables machine learning models to be trained onmultiple datasets without compromising user privacy, making it particularlywell-suited for addressing the challenges of survival data and large-scalesurvival applications. Despite significant developments in federated learningfor classification and regression, many directions remain unexplored in thecontext of survival analysis. In this work, we propose an extension of theFederated Survival Forest algorithm, called FedSurF++. This federated ensemblemethod constructs random survival forests in heterogeneous federations.Specifically, we investigate several new tree sampling methods from clientforests and compare the results with state-of-the-art survival models based onneural networks. The key advantage of FedSurF++ is its ability to achievecomparable performance to existing methods while requiring only a singlecommunication round to complete. The extensive empirical investigation resultsin a significant improvement from the algorithmic and privacy preservationperspectives, making the original FedSurF algorithm more efficient, robust, andprivate. We also present results on two real-world datasets demonstrating thesuccess of FedSurF++ in real-world healthcare studies. Our results underscorethe potential of FedSurF++ to improve the scalability and effectiveness ofsurvival analysis in distributed settings while preserving user privacy.

Scaling Survival Analysis in Healthcare with Federated Survival Forests: A Comparative Study on Heart Failure and Breast Cancer Genomics