Anatomica: Localized Control over Geometric and Topological Properties for Anatomical Diffusion Models
This work addresses the need for controlled generation of anatomical data for applications like virtual trials, offering a domain-specific solution.
The authors tackled the problem of generating anatomical voxel maps with precise control over geometric and topological properties, resulting in a framework that enables localized constraints on features like size, shape, and connectivity for synthetic dataset creation.
We present Anatomica: an inference-time framework for generating multi-class anatomical voxel maps with localized geo-topological control. During generation, we use cuboidal control domains of varying dimensionality, location, and shape to slice out relevant substructures. These local substructures are used to compute differentiable penalty functions that steer the sample towards target constraints. We control geometric features such as size, shape, and position through voxel-wise moments, while topological features such as connected components, loops, and voids are enforced through persistent homology. Lastly, we implement Anatomica for latent diffusion models, where neural field decoders partially extract substructures, enabling the efficient control of anatomical properties. Anatomica applies flexibly across diverse anatomical systems, composing constraints to control complex structures over arbitrary dimensions and coordinate systems, thereby enabling the rational design of synthetic datasets for virtual trials or machine learning workflows.