Neuromorphologicaly-preserving Volumetric data encoding using VQ-VAE

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 $0.825\%$ 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.

References in corpus (7)

• Adam: A Method for Stochastic Optimization
• Neural Discrete Representation Learning
• NiftyNet: a deep-learning platform for medical imaging
• Self-training with Noisy Student improves ImageNet classification
• Checkerboard artifact free sub-pixel convolution: A note on sub-pixel convolution, resize convolution and convolution resize
• Generative Image Translation for Data Augmentation of Bone Lesion Pathology
• FMRI data augmentation via synthesis

Cited by in corpus (4)

• FakeNews: GAN-based generation of realistic 3D volumetric data -- A systematic review and taxonomy
• Generative Aging of Brain Images with Diffeomorphic Registration
• Bayesian Image Reconstruction using Deep Generative Models
• Latent Space Analysis of VAE and Intro-VAE applied to 3-dimensional MR Brain Volumes of Multiple Sclerosis, Leukoencephalopathy, and Healthy Patients