Spinverse: Differentiable Physics for Permeability-Aware Microstructure Reconstruction from Diffusion MRI
This work addresses the problem of reconstructing explicit microstructural interfaces from dMRI, which is crucial for understanding tissue microstructure in medical imaging, particularly for researchers and clinicians studying neurological conditions.
This paper introduces Spinverse, a method for reconstructing microstructural interfaces from diffusion MRI (dMRI) by inverting dMRI measurements through a differentiable Bloch-Torrey simulator. It represents tissue on a tetrahedral grid and learns face permeabilities, which allows microstructural boundaries to emerge without pre-defined topology. The method successfully reconstructs diverse geometries on synthetic voxel meshes, highlighting the importance of sequence scheduling and regularization for accurate and structurally valid boundaries.
Diffusion MRI (dMRI) is sensitive to microstructural barriers, yet most existing methods either assume impermeable boundaries or estimate voxel-level parameters without recovering explicit interfaces. We present Spinverse, a permeability-aware reconstruction method that inverts dMRI measurements through a fully differentiable Bloch-Torrey simulator. Spinverse represents tissue on a fixed tetrahedral grid and treats each interior face permeability as a learnable parameter; low-permeability faces act as diffusion barriers, so microstructural boundaries whose topology is not fixed a priori (up to the resolution of the ambient mesh) emerge without changing mesh connectivity or vertex positions. Given a target signal, we optimize face permeabilities by backpropagating a signal-matching loss through the PDE forward model, and recover an interface by thresholding the learned permeability field. To mitigate the ill-posedness of permeability inversion, we use mesh-based geometric priors; to avoid local minima, we use a staged multi-sequence optimization curriculum. Across a collection of synthetic voxel meshes, Spinverse reconstructs diverse geometries and demonstrates that sequence scheduling and regularization are critical to avoid outline-only solutions while improving both boundary accuracy and structural validity.