Federated Timeline Synthesis: Scalable and Private Methodology For Model Training and Deployment
This addresses privacy and scalability challenges in healthcare AI by enabling model training across institutions without sharing raw data, though it is incremental as it builds on federated learning and generative models.
The paper tackles the problem of training generative foundation models on distributed timeseries data like electronic health records by introducing Federated Timeline Synthesis (FTS), which uses local transformers and a global generator to synthesize patient trajectories, achieving performance comparable to real data on five clinical prediction tasks.
We present Federated Timeline Synthesis (FTS), a novel framework for training generative foundation models across distributed timeseries data applied to electronic health records (EHR). At its core, FTS represents patient history as tokenized Patient Health Timelines (PHTs), language-agnostic sequences encoding temporal, categorical, and continuous clinical information. Each institution trains an autoregressive transformer on its local PHTs and transmits only model weights to a central server. The server uses the generators to synthesize a large corpus of trajectories and train a Global Generator (GG), enabling zero-shot inference via Monte Carlo simulation of future PHTs. We evaluate FTS on five clinically meaningful prediction tasks using MIMIC-IV data, showing that models trained on synthetic data generated by GG perform comparably to those trained on real data. FTS offers strong privacy guarantees, scalability across institutions, and extensibility to diverse prediction and simulation tasks especially in healthcare, including counterfactual inference, early warning detection, and synthetic trial design.