TRELLIS-Enhanced Surface Features for Comprehensive Intracranial Aneurysm Analysis

Intracranial aneurysms pose a significant clinical risk yet are difficult to detect, delineate, and model due to limited annotated 3D data. We propose a cross-domain feature-transfer approach that leverages the latent geometric embeddings learned by TRELLIS, a generative model trained on large-scale non-medical 3D datasets, to augment neural networks for aneurysm analysis. By replacing conventional point normals or mesh descriptors with TRELLIS surface features, we systematically enhance three downstream tasks: (i) classifying aneurysms versus healthy vessels in the Intra3D dataset, (ii) segmenting aneurysm and vessel regions on 3D meshes, and (iii) predicting time-evolving blood-flow fields using a graph neural network on the AnX-plore dataset. Our experiments show that the inclusion of these features yields strong gains in accuracy, F1-score, and segmentation quality over state-of-the-art baselines, and reduces simulation error by 15%. These results illustrate the broader potential of transferring 3D representations from general-purpose generative models to specialized medical tasks.1. Introduction Intracranial aneurysms represent a significant and often silent threat to human health. These pathologies typically remain asymptomatic until the moment of rupture, a catastrophic event associated with high rates of morbidity and mortality. Consequently, the early and accurate detection of un-ruptured aneurysms is crucial for preventive clinical intervention. Advanced imaging modalities, particularly Magnetic Resonance Angiography (MRA), have become instrumental in this endeavor, enabling the reconstruction of detailed 3D models of the brain's vascular network to facilitate the identification of aneurysms.