ItemConference Paper

FedDiff-Health: A Privacy-Preserving Generative Framework for Collaborative Hospital Readmission Prediction

Title
FedDiff-Health: A Privacy-Preserving Generative Framework for Collaborative Hospital Readmission Prediction
Creator
Santhrupth, B.C.; Bhoomika, G.R.; Jeevaraj, R.; Banakar, Shivaraj Veerappa; Shreyas, L.; Rakesh, V.S.
Description
Hospital readmission prediction encounters three challenges: data siloing across hospitals due to format incompatibilities, stringent privacy constraints, and the rarity of readmission events. We propose P-Fed-Diffusion, the first framework that enables collaboration across hospitals while keeping patients' data private. Our method automatically aligns heterogeneous data schemas without human intervention, using large language models. Then, we apply conditional diffusion models within a federated learning framework to generate synthetic data for rare readmission events. The framework incorporates formal privacy guarantees via differential privacy. We achieve a dramatic improvement over state-of-the-art methods: while the best prior method achieves 2% recall, we achieve 64% recall-32x improvement, meaning that the method finds over 1,000 additional high-risk patients per hospital annually. Our work opens up a new direction for privacy-preserving collaborative AI across hospitals. 2025 IEEE.
Source
4th International Conference on Applied Artificial Intelligence and Computing, ICAAIC 2025;pp.1539-1544
Date
01-01-2025
Publisher
Institute of Electrical and Electronics Engineers Inc.
Subject
Differential Privacy; Electronic Health Records; Federated Learning; Hospital Readmission Prediction; Schema Harmonization
Coverage
Santhrupth B.C., Christ University, Department of AI, Ml&ds, Karnataka, Bangalore, India; Bhoomika G.R., Christ University, Department of Cse, Karnataka, Bangalore, India; Jeevaraj R., Global Academy of Technology, Department of Ise, Karnataka, Bangalore, India; Banakar S.V., Global Academy of Technology, Department of Ise, Karnataka, Bangalore, India; Shreyas L., Global Academy of Technology, Department of Ise, Karnataka, Bangalore, India; Rakesh V.S., Cambridge Institute of Technology, Department of Cse, Karnataka, Bangalore, India
Rights
Restricted Access; Hardcopy may be available in the library
Relation
ISBN: 979-833156587-9;
Format
online
Language
English
Type
Conference paper

Citation

Santhrupth, B.C.; Bhoomika, G.R.; Jeevaraj, R.; Banakar, Shivaraj Veerappa; Shreyas, L.; Rakesh, V.S., “FedDiff-Health: A Privacy-Preserving Generative Framework for Collaborative Hospital Readmission Prediction,” CHRIST (Deemed To Be University) Institutional Repository, accessed June 18, 2026, https://archives.christuniversity.in/items/show/25888.

Output Formats