PrIntMesh: Precise Intersection Surfaces for 3D Organ Mesh Reconstruction
This provides a data-efficient solution for clinical use by improving reconstruction of shared interfaces and structural consistency in organs like the heart, hippocampus, and lungs.
The paper tackled the problem of anatomically implausible 3D organ mesh reconstructions by introducing PrIntMesh, a template-based framework that jointly deforms interconnected substructures to match patient-specific anatomy, achieving high geometric accuracy and robust performance with limited or noisy data.
Human organs are composed of interconnected substructures whose geometry and spatial relationships constrain one another. Yet, most deep-learning approaches treat these parts independently, producing anatomically implausible reconstructions. We introduce PrIntMesh, a template-based, topology-preserving framework that reconstructs organs as unified systems. Starting from a connected template, PrIntMesh jointly deforms all substructures to match patient-specific anatomy, while explicitly preserving internal boundaries and enforcing smooth, artifact-free surfaces. We demonstrate its effectiveness on the heart, hippocampus, and lungs, achieving high geometric accuracy, correct topology, and robust performance even with limited or noisy training data. Compared to voxel- and surface-based methods, PrIntMesh better reconstructs shared interfaces, maintains structural consistency, and provides a data-efficient solution suitable for clinical use.