Temporal Consistency Objectives Regularize the Learning of Disentangled Representations

  title={Temporal Consistency Objectives Regularize the Learning of Disentangled Representations},
  author={Gabriele Valvano and Agisilaos Chartsias and Andrea Leo and Sotirios A. Tsaftaris},
There has been an increasing focus in learning interpretable feature representations, particularly in applications such as medical image analysis that require explainability, whilst relying less on annotated data (since annotations can be tedious and costly. [...] Key Method This forces the anatomical decomposition to be consistent with the temporal cardiac contraction in cine MRI and to have semantic meaning with less need for annotations. We demonstrate that using this regularization, we achieve competitive…Expand
Self-supervised Multi-scale Consistency for Weakly Supervised Segmentation Learning
A novel self-supervised multiscale consistency loss is introduced, which, coupled with an attention mechanism, encourages the segmentor to learn multi-scale relationships between objects and improves performance. Expand
Combined Spiral Transformation and Model-Driven Multi-Modal Deep Learning Scheme for Automatic Prediction of TP53 Mutation in Pancreatic Cancer
The experimental results show that the proposed model gives the desired performance in predicting TP53 mutation in pancreatic cancer, providing a new approach for noninvasive gene prediction. Expand
A Tutorial on Learning Disentangled Representations in the Imaging Domain
This tutorial paper offers an overview of the disentangled representation learning, its building blocks and criteria, and discusses applications in computer vision and medical imaging, and presents the identified opportunities for the integration of recent machine learning advances into disentanglement. Expand
FOAL: Fast Online Adaptive Learning for Cardiac Motion Estimation
A novel fast online adaptive learning (FOAL) framework: an online gradient descent based optimizer that is optimized by a meta-learner that enables the online optimizer to perform a fast and robust adaptation. Expand


Disentangled representation learning in cardiac image analysis
Spatial Decomposition Network (SDNet) is proposed, which factorises 2D medical images into spatial anatomical factors and non-spatial modality factors and is ideally suited for several medical image analysis tasks, such as semi-supervised segmentation, multi-task segmentation and regression, and image-toimage synthesis. Expand
Recurrent neural networks for aortic image sequence segmentation with sparse annotations
Experiments demonstrate that the proposed image sequence segmentation algorithm by combining a fully convolutional network with a recurrent neural network significantly improves both accuracy and temporal smoothness of segmentation, compared to a baseline method that utilises spatial information only. Expand
Learning to Decompose and Disentangle Representations for Video Prediction
The Decompositional Disentangled Predictive Auto-Encoder (DDPAE) is proposed, a framework that combines structured probabilistic models and deep networks to automatically decompose the high-dimensional video that the authors aim to predict into components, and disentangle each component to have low-dimensional temporal dynamics that are easier to predict. Expand
U-Net: Convolutional Networks for Biomedical Image Segmentation
It is shown that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks. Expand
Deep Learning Techniques for Automatic MRI Cardiac Multi-Structures Segmentation and Diagnosis: Is the Problem Solved?
How far state-of-the-art deep learning methods can go at assessing CMRI, i.e., segmenting the myocardium and the two ventricles as well as classifying pathologies is measured, to open the door to highly accurate and fully automatic analysis of cardiac CMRI. Expand
Unsupervised Deformable Registration for Multi-Modal Images via Disentangled Representations
This paper proposes a fully unsupervised multi-modal deformable image registration method (UMDIR), which does not require any ground truth deformation fields or any aligned multi- modal image pairs during training, and achieves competitive performance against other methods at substantially reduced computation time. Expand
Joint Learning of Motion Estimation and Segmentation for Cardiac MR Image Sequences
Experimental results show that the joint learning of both tasks is complementary and the proposed models outperform the competing methods significantly in terms of accuracy and speed. Expand
InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
Experiments show that InfoGAN learns interpretable representations that are competitive with representations learned by existing fully supervised methods. Expand
Representation Learning: A Review and New Perspectives
Recent work in the area of unsupervised feature learning and deep learning is reviewed, covering advances in probabilistic models, autoencoders, manifold learning, and deep networks. Expand
Diverse Image-to-Image Translation via Disentangled Representations
This work presents an approach based on disentangled representation for producing diverse outputs without paired training images, and proposes to embed images onto two spaces: a domain-invariant content space capturing shared information across domains and adomain-specific attribute space. Expand