Sparse Representation Learning for Vessels

TL;DR

This paper introduces VAEsselSparse, a novel sparse convolutional model that efficiently encodes entire organ-level vascular networks at high resolution, enabling accurate reconstruction, classification, and realistic vessel synthesis.

Contribution

The paper presents a new sparse encoder-decoder model that captures organ-level vascular structures at sub-millimeter resolution with high compression and clinical relevance.

Findings

Achieves 8x8x8 spatial compression of vascular networks.

Outperforms dense models and previous methods in reconstruction accuracy.

Latent space effectively supports vessel classification and realistic synthesis.

Abstract

Analyzing human vasculature and vessel-like, tubular structures, such as airways, is crucial for disease diagnosis and treatment. Current methods often rely on small sub-regions or simplified tree-like structures, rendering analysis of entire organ-level networks at clinical resolution computationally challenging. To this end, we propose VAEsselSparse, an efficient encoder-decoder model to obtain a meaningful yet compact representation of the entire organ-level vascular network at sub-millimeter resolution. VAEsselSparse leverages the inherent sparsity of 3D vascular structures via sparse convolutions and attention mechanisms, achieving substantial spatial compression rates of 8 x 8 x 8. We demonstrate superior reconstruction performance compared to dense counterparts and previous methods. Importantly, the resulting latent space retains clinically relevant discriminative features readily usable for classification tasks, such as aneurysm/stenosis or subvariants of the circle of Willis. Moreover, the compact latent space of VAEsselSparse serves as an effective representation for learning vessel-specific priors through generative models, enabling the synthesis of realistic vasculature.