VecHeart: Holistic Four-Chamber Cardiac Anatomy Modeling via Hybrid VecSets

2026-04-21

AI Summary

Existing methods struggle to model the complex anatomical dependencies among the four cardiac chambers and fail to reconstruct complete dynamic 3D+t structures from sparse or missing observations. This work proposes VecHeart, a novel framework that introduces a hybrid part-based Transformer architecture, combining part-specific learnable queries with interleaved attention mechanisms to enable collaborative multi-chamber modeling. By incorporating an anatomical completion mask and a modality alignment strategy, VecHeart supports high-fidelity inference of missing components from incomplete data and generates temporally consistent 3D+t mesh sequences. The method achieves state-of-the-art performance across multiple challenging scenarios, significantly outperforming existing approaches in both reconstruction accuracy and temporal coherence.

Technology Category

• Computer Vision: 3D Computer Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Multimodal Learning

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Accurate cardiac anatomy modeling requires the model to be able to handle intricate interrelations among structures. In this paper, we propose VecHeart, a unified framework for holistic reconstruction and generation of four-chamber cardiac structures. To overcome the limitations of current feed-forward implicit methods, specifically their restriction to single-object modeling and their neglect of inter-part correlations, we introduce Hybrid Part Transformer, which leverages part-specific learnable queries and interleaved attention to capture complex inter-chamber dependencies. Furthermore, we propose Anatomical Completion Masking and Modality Alignment strategies, enabling the model to infer complete four-chamber structures from partial, sparse, or noisy observations, even when certain anatomical parts are entirely missing. VecHeart also seamlessly extends to 3D+t dynamic mesh sequence generation, demonstrating exceptional versatility. Experiments show that our method achieves state-of-the-art performance, maintaining high-fidelity reconstruction across diverse challenging scenarios. Code will be released.

Problem

• cardiac anatomy modeling
• four-chamber heart
• inter-part correlations
• partial observations
• holistic reconstruction

Innovation

• Hybrid Part Transformer
• Anatomical Completion Masking
• Modality Alignment
• Four-chamber cardiac modeling
• Implicit neural representation