Current browse context:
eess
Change to browse by:
References & Citations
Electrical Engineering and Systems Science > Image and Video Processing
Title: Neuromorphologicaly-preserving Volumetric data encoding using VQ-VAE
(Submitted on 13 Feb 2020)
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.
Submission history
From: Petru-Daniel Tudosiu [view email][v1] Thu, 13 Feb 2020 18:18:51 GMT (5031kb,D)
Link back to: arXiv, form interface, contact.