SuperFormer: Volumetric Transformer Architectures for MRI Super-Resolution
This work addresses MRI super-resolution for medical imaging, offering an incremental improvement by adapting existing transformer architectures to 3D volumetric data.
The paper tackles MRI super-resolution by extending Swin Transformers to 3D medical data, proposing a volumetric transformer with 3D relative positional encoding and multi-domain fusion, achieving superior results over 3D CNN-based methods on the Human Connectome Project dataset.
This paper presents a novel framework for processing volumetric medical information using Visual Transformers (ViTs). First, We extend the state-of-the-art Swin Transformer model to the 3D medical domain. Second, we propose a new approach for processing volumetric information and encoding position in ViTs for 3D applications. We instantiate the proposed framework and present SuperFormer, a volumetric transformer-based approach for Magnetic Resonance Imaging (MRI) Super-Resolution. Our method leverages the 3D information of the MRI domain and uses a local self-attention mechanism with a 3D relative positional encoding to recover anatomical details. In addition, our approach takes advantage of multi-domain information from volume and feature domains and fuses them to reconstruct the High-Resolution MRI. We perform an extensive validation on the Human Connectome Project dataset and demonstrate the superiority of volumetric transformers over 3D CNN-based methods. Our code and pretrained models are available at https://github.com/BCV-Uniandes/SuperFormer.