English

Neuromorphologicaly-preserving Volumetric data encoding using VQ-VAE

Image and Video Processing 2020-02-14 v1 Computer Vision and Pattern Recognition Quantitative Methods

Abstract

The increasing efficiency and compactness of deep learning architectures, together with hardware improvements, have enabled the complex and high-dimensional modelling of medical volumetric data at higher resolutions. Recently, Vector-Quantised Variational Autoencoders (VQ-VAE) have been proposed as an efficient generative unsupervised learning approach that can encode images to a small percentage of their initial size, while preserving their decoded fidelity. Here, we show a VQ-VAE inspired network can efficiently encode a full-resolution 3D brain volume, compressing the data to 0.825%0.825\% of the original size while maintaining image fidelity, and significantly outperforming the previous state-of-the-art. We then demonstrate that VQ-VAE decoded images preserve the morphological characteristics of the original data through voxel-based morphology and segmentation experiments. Lastly, we show that such models can be pre-trained and then fine-tuned on different datasets without the introduction of bias.

Keywords

Cite

@article{arxiv.2002.05692,
  title  = {Neuromorphologicaly-preserving Volumetric data encoding using VQ-VAE},
  author = {Petru-Daniel Tudosiu and Thomas Varsavsky and Richard Shaw and Mark Graham and Parashkev Nachev and Sebastien Ourselin and Carole H. Sudre and M. Jorge Cardoso},
  journal= {arXiv preprint arXiv:2002.05692},
  year   = {2020}
}
R2 v1 2026-06-23T13:41:11.717Z