Özgün Turgut

Clinical histories are fundamentally irregular and multimodal, yet foundation models have treated them as static or uniformly discrete—until a generative variational transformer trained on 559 million events unified continuous-time forecasting, multi-modal health records, and counterfactual intervention simulation.

Multi-view Echo data can be used to train ECG encoders that are 18x smaller yet outperform larger models at predicting cardiac morphology.