Prior-Fitted Functional Flow: In-Context Generative Models for Pharmacokinetics
César Ojeda, Niklas Hartung, Wilhelm Huisinga, Tim Jahn, Purity Kamene Kavwele + 4 more
TLDR
Prior-Fitted Functional Flows is a generative model for pharmacokinetics, enabling zero-shot population synthesis and individual forecasting.
Key contributions
- Introduces Prior-Fitted Functional Flows, a generative model for pharmacokinetics.
- Enables zero-shot population synthesis and individual patient forecasting without manual tuning.
- Learns functional vector fields from sparse, irregular study population data.
- Achieves state-of-the-art predictive accuracy on extensive real-world datasets.
Why it matters
This paper introduces a novel generative model for pharmacokinetics, crucial for drug development and personalized medicine. It enables zero-shot forecasting and population synthesis, accelerating drug discovery and optimizing patient treatments. Handling sparse data with calibrated uncertainty is a key advancement.
Original Abstract
We introduce Prior-Fitted Functional Flows, a generative foundation model for pharmacokinetics that enables zero-shot population synthesis and individual forecasting without manual parameter tuning. We learn functional vector fields, explicitly conditioned on the sparse, irregular data of an entire study population. This enables the generation of coherent virtual cohorts as well as forecasting of partially observed patient trajectories with calibrated uncertainty. We construct a new open-access literature corpus to inform our priors, and demonstrate state-of-the-art predictive accuracy on extensive real-world datasets.
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.