Search papers, labs, and topics across Lattice.
This study develops a multimodal generative framework for neuroimaging data, specifically focusing on structural and functional MRI features by evaluating various encoding strategies and generative model architectures. The researchers employ a novel modality-aware graph encoding approach that significantly enhances the performance of latent space representations compared to traditional vectorized methods. The proposed multimodal graph VAE (gMMVAE) demonstrates superior generation fidelity and reconstruction quality, establishing itself as a powerful tool for multimodal neuroimaging analysis.
Modality-aware graph encoding in generative models can dramatically enhance the fidelity and efficiency of neuroimaging feature analysis.
While generative models enable encoding of complex neuroimaging data for feature generation and reconstruction, developing optimal architectural frameworks with appropriate encoding and latent space processes is crucial for studying structural and functional properties of the brain. We design a multimodal generative framework for structural and functional magnetic resonance imaging (MRI) features through systematic evaluation of encoding strategies, latent multimodal fusion, and generative model selection. Using structural gray matter volume (GMV) and static functional network connectivity (sFNC) features from a large neuroimaging dataset, we analyze generative frameworks involving variational autoencoders (VAEs), transformers, generative adversarial networks (GANs), and diffusion models. Architectures that employ modality-aware graph encoding of functional connectivity into a lower-dimensional latent space outperform vectorized encoders or direct data space approaches. The proposed multimodal graph VAE (gMMVAE) surpasses alternative generative variants across multiple metrics for generation fidelity, reconstruction quality, efficiency, and latent space discriminability, highlighting its potential for robust multimodal neuroimaging analysis.
