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

TL;DR

VecHeart is a comprehensive framework that models four-chamber cardiac anatomy by capturing interrelations and handling partial or noisy data, extending to dynamic mesh sequences with state-of-the-art accuracy.

Contribution

Introduces Hybrid Part Transformer and novel strategies for holistic cardiac structure modeling from partial data, surpassing limitations of existing implicit methods.

Findings

Achieves state-of-the-art reconstruction accuracy.

Effectively infers complete structures from partial or noisy observations.

Extends to dynamic 3D+t mesh sequence generation.

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.