High-Fidelity Medical Shape Generation via Skeletal Latent Diffusion

TL;DR

This paper introduces a skeletal latent diffusion framework for high-fidelity medical shape generation, leveraging structural priors and a large-scale dataset to improve accuracy and efficiency in modeling complex anatomical structures.

Contribution

It presents a novel skeletal latent diffusion method with a shape auto-encoder and a large-scale MedSDF dataset for improved medical shape generation.

Findings

Superior reconstruction quality on MedSDF and vessel datasets

Higher computational efficiency than existing methods

Effective modeling of complex anatomical structures

Abstract

Anatomy shape modeling is a fundamental problem in medical data analysis. However, the geometric complexity and topological variability of anatomical structures pose significant challenges to accurate anatomical shape generation. In this work, we propose a skeletal latent diffusion framework that explicitly incorporates structural priors for efficient and high-fidelity medical shape generation. We introduce a shape auto-encoder in which the encoder captures global geometric information through a differentiable skeletonization module and aggregates local surface features into shape latents, while the decoder predicts the corresponding implicit fields over sparsely sampled coordinates. New shapes are generated via a latent-space diffusion model, followed by neural implicit decoding and mesh extraction. To address the limited availability of medical shape data, we construct a large-scale dataset, \textit{MedSDF}, comprising surface point clouds and corresponding signed distance fields across multiple anatomical categories. Extensive experiments on MedSDF and vessel datasets demonstrate that the proposed method achieves superior reconstruction and generation quality while maintaining a higher computational efficiency compared with existing approaches. Code is available at: https://github.com/wlsdzyzl/meshage.