🤖 AI Summary
To address the high computational cost and poor scalability of template generation in group-wise brain MRI registration, this paper proposes a real-time template generation framework based on a dual-VAE architecture. Methodologically, it pioneers zero-shot template synthesis by exploiting the arithmetic additivity of VAE latent spaces; introduces a Displacement Inversion Module (DIM) to ensure template unbiasedness; and designs a Subject-to-Template Alignment Module (STAM) to enhance structural consistency. Evaluated on OASIS and ADNI datasets, the framework generates high-quality, unbiased, and highly aligned group templates within seconds. It significantly outperforms state-of-the-art methods in both registration accuracy and template unbiasedness, while achieving over two orders-of-magnitude speedup in runtime. To our knowledge, this is the first approach enabling scalable, real-time, and training-free group template construction.
📝 Abstract
Template generation is a critical step in groupwise image registration, which involves aligning a group of subjects into a common space. While existing methods can generate high-quality template images, they often incur substantial time costs or are limited by fixed group scales. In this paper, we present InstantGroup, an efficient groupwise template generation framework based on variational autoencoder (VAE) models that leverage latent representations' arithmetic properties, enabling scalability to groups of any size. InstantGroup features a Dual VAEs backbone with shared-weight twin networks to handle pairs of inputs and incorporates a Displacement Inversion Module (DIM) to maintain template unbiasedness and a Subject-Template Alignment Module (STAM) to improve template quality and registration accuracy. Experiments on 3D brain MRI scans from the OASIS and ADNI datasets reveal that InstantGroup dramatically reduces runtime, generating templates within seconds for various group sizes while maintaining superior performance compared to state-of-the-art baselines on quantitative metrics, including unbiasedness and registration accuracy.
