Two-Stage Generative Model for Intracranial Aneurysm Meshes with Morphological Marker Conditioning

2025-05-15

AI Summary

Existing intracranial aneurysm (IA) generation methods struggle to simultaneously ensure morphological realism and physiological plausibility, particularly neglecting the geometric relationship between the aneurysmal sac and parent vessel, and lacking controllable synthesis guided by clinical morphological indices (e.g., aspect ratio, irregularity). This paper proposes the first morphology-parameter-controllable IA mesh generation model. It employs a two-stage variational autoencoder (VAE) framework: the first stage encodes sac geometry via graph-based harmonic deformation (GHD) while regularizing deformation energy; the second stage conditions on GHD features to jointly model parent vessel centerline evolution and cross-sectional dynamics, enabling end-to-end, clinically index-driven generation. The model significantly improves geometric fidelity, vascular connectivity plausibility, and morphological controllability, facilitating high-fidelity hemodynamic simulations even under data scarcity. The code is publicly available.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Natural Language Processing: Generation

Application Category

• Web Mining and Content Analysis: Web data generation and simulation
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

A generative model for the mesh geometry of intracranial aneurysms (IA) is crucial for training networks to predict blood flow forces in real time, which is a key factor affecting disease progression. This need is necessitated by the absence of a large IA image datasets. Existing shape generation methods struggle to capture realistic IA features and ignore the relationship between IA pouches and parent vessels, limiting physiological realism and their generation cannot be controlled to have specific morphological measurements. We propose AneuG, a two-stage Variational Autoencoder (VAE)-based IA mesh generator. In the first stage, AneuG generates low-dimensional Graph Harmonic Deformation (GHD) tokens to encode and reconstruct aneurysm pouch shapes, constrained to morphing energy statistics truths. GHD enables more accurate shape encoding than alternatives. In the second stage, AneuG generates parent vessels conditioned on GHD tokens, by generating vascular centreline and propagating the cross-section. AneuG's IA shape generation can further be conditioned to have specific clinically relevant morphological measurements. This is useful for studies to understand shape variations represented by clinical measurements, and for flow simulation studies to understand effects of specific clinical shape parameters on fluid dynamics. Source code and implementation details are available at https://github.com/anonymousaneug/AneuG.

Problem

• Generates realistic intracranial aneurysm meshes lacking large datasets
• Captures aneurysm-parent vessel relationships for physiological realism
• Controls mesh generation with clinical morphological measurements

Innovation

• Two-stage VAE for aneurysm mesh generation
• Graph Harmonic Deformation for shape encoding
• Conditional generation with clinical measurements