Brain fluid flow mapped with physics-informed neural modeling
PI-NOMT: Physics-Informed Neural Optimal Mass Transport for Brain Fluid Dynamics
Machine LearningArtificial Intelligence
Summary
It is hard to figure out how fluids move inside the brain just by looking at snapshots taken over time. The authors propose a new method that uses knowledge of physical laws and a special kind of neural network to guess the hidden flow and sources of these fluids. Their method not only matches the observed fluid concentrations but also recovers the actual flow speeds and directions. They tested this on both simulated data and real brain scans of rats and found their method to be accurate and physically consistent.
What this means in practice
- •For medical imaging engineers: Estimate hidden brain fluid flow fields from tracer MRI sequences for improved diagnostic analysis.
- •For computational neuroscientists: Model and analyze transport mechanisms of substances in brain tissue using continuous neural representations guided by physics.
Authors
Mehmet Emin Acar, Vahit Bugra Yesilkaynak, Helene Benveniste, Gozde Unal
Abstract
Recovering hidden transport mechanisms from sparse spatiotemporal observations is a fundamental inverse problem in scientific machine learning. In brain tracer imaging, dynamic contrast-enhanced MRI (DCE-MRI) provides time-resolved measurements of tracer concentration, while the underlying velocity and source mechanisms governing tracer propagation remain unobserved. We formulate this problem as physics-informed latent-state inference, in which the transport field itself is the primary object of inference rather than an auxiliary variable used only to reconstruct observed densities. We propose Physics-Informed Neural Optimal Mass Transport (PI-NOMT), a framework that represents density, velocity, and source as continuous neural fields and combines a continuous neural density teacher, recursive differentiable advection--diffusion--source rollout, unbalanced optimal-transport regularization, and governing-equation supervision. Physical laws act as structural priors that constrain the space of admissible transport mechanisms, while observed tracer dynamics provide evidence for estimating the latent transport state. We evaluate PI-NOMT on a synthetic benchmark with known ground-truth transport and on DCE-MRI sequences from nine control rats. On the synthetic benchmark, PI-NOMT accurately recovers the prescribed velocity field, including its magnitude, direction, and integrated trajectories, rather than merely reconstructing endpoint densities. Across the nine rat datasets, the framework yields sub-percent local endpoint error, consistent physical speed scales, and low post-training PDE and incompressibility residuals. These results support physics-informed latent-state inference as a general framework for recovering hidden transport mechanisms from observed dynamic scalar fields.