VesselVAE: Recursive Variational Autoencoders for 3D Blood Vessel Synthesis

TL;DR

VesselVAE introduces a recursive variational autoencoder that captures the hierarchical structure of blood vessels, enabling the generation of diverse and realistic 3D vessel geometries for medical applications.

Contribution

This work is the first to apply recursive variational autoencoders for synthesizing blood vessel geometries, capturing complex structural variations effectively.

Findings

High similarity between synthetic and real vessel data (radius 0.97, length 0.95, tortuosity 0.96)

Generates diverse and anatomically plausible 3D vessel models

Enables applications in medical training and hemodynamic simulations

Abstract

We present a data-driven generative framework for synthesizing blood vessel 3D geometry. This is a challenging task due to the complexity of vascular systems, which are highly variating in shape, size, and structure. Existing model-based methods provide some degree of control and variation in the structures produced, but fail to capture the diversity of actual anatomical data. We developed VesselVAE, a recursive variational Neural Network that fully exploits the hierarchical organization of the vessel and learns a low-dimensional manifold encoding branch connectivity along with geometry features describing the target surface. After training, the VesselVAE latent space can be sampled to generate new vessel geometries. To the best of our knowledge, this work is the first to utilize this technique for synthesizing blood vessels. We achieve similarities of synthetic and real data for radius (.97), length (.95), and tortuosity (.96). By leveraging the power of deep neural networks, we generate 3D models of blood vessels that are both accurate and diverse, which is crucial for medical and surgical training, hemodynamic simulations, and many other purposes.