Thomas Joyce

Bio: Thomas Joyce is an academic researcher from University of Edinburgh. The author has contributed to research in topics: Segmentation & Metaheuristic. The author has an hindex of 11, co-authored 18 publications receiving 771 citations. Previous affiliations of Thomas Joyce include ETH Zurich & University of Zurich.

Learning motion manifolds with convolutional autoencoders

Abstract: We present a technique for learning a manifold of human motion data using Convolutional Autoencoders. Our approach is capable of learning a manifold on the complete CMU database of human motion. This manifold can be treated as a prior probability distribution over human motion data, which has many applications in animation research, including projecting invalid or corrupt motion onto the manifold for removing error, computing similarity between motions using geodesic distance along the manifold, and interpolation of motion along the manifold for avoiding blending artefacts.

Multimodal MR Synthesis via Modality-Invariant Latent Representation

Abstract: We propose a multi-input multi-output fully convolutional neural network model for MRI synthesis. The model is robust to missing data, as it benefits from, but does not require, additional input modalities. The model is trained end-to-end, and learns to embed all input modalities into a shared modality-invariant latent space. These latent representations are then combined into a single fused representation, which is transformed into the target output modality with a learnt decoder. We avoid the need for curriculum learning by exploiting the fact that the various input modalities are highly correlated. We also show that by incorporating information from segmentation masks the model can both decrease its error and generate data with synthetic lesions. We evaluate our model on the ISLES and BRATS data sets and demonstrate statistically significant improvements over state-of-the-art methods for single input tasks. This improvement increases further when multiple input modalities are used, demonstrating the benefits of learning a common latent space, again resulting in a statistically significant improvement over the current best method. Finally, we demonstrate our approach on non skull-stripped brain images, producing a statistically significant improvement over the previous best method. Code is made publicly available at https://github.com/agis85/multimodal_brain_synthesis .

Adversarial Image Synthesis for Unpaired Multi-modal Cardiac Data

Abstract: This paper demonstrates the potential for synthesis of medical images in one modality (e.g. MR) from images in another (e.g. CT) using a CycleGAN [24] architecture. The synthesis can be learned from unpaired images, and applied directly to expand the quantity of available training data for a given task. We demonstrate the application of this approach in synthesising cardiac MR images from CT images, using a dataset of MR and CT images coming from different patients. Since there can be no direct evaluation of the synthetic images, as no ground truth images exist, we demonstrate their utility by leveraging our synthetic data to achieve improved results in segmentation. Specifically, we show that training on both real and synthetic data increases accuracy by 15% compared to real data. Additionally, our synthetic data is of sufficient quality to be used alone to train a segmentation neural network, that achieves 95% of the accuracy of the same model trained on real data.

A Review of No Free Lunch Theorems, and Their Implications for Metaheuristic Optimisation

Abstract: The No Free Lunch Theorem states that, averaged over all optimisation problems, all non-resampling optimisation algorithms perform equally well. In order to explain the relevance of these theorems for metaheuristic optimisation, we present a detailed discussion on the No Free Lunch Theorem, and various extensions including some which have not appeared in the literature so far. We then show that understanding the No Free Lunch theorems brings us to a position where we can ask about the specific dynamics of an optimisation algorithm, and how those dynamics relate to the properties of optimisation problems.

A deep learning framework for character motion synthesis and editing

Abstract: We present a framework to synthesize character movements based on high level parameters, such that the produced movements respect the manifold of human motion, trained on a large motion capture dataset. The learned motion manifold, which is represented by the hidden units of a convolutional autoencoder, represents motion data in sparse components which can be combined to produce a wide range of complex movements. To map from high level parameters to the motion manifold, we stack a deep feedforward neural network on top of the trained autoencoder. This network is trained to produce realistic motion sequences from parameters such as a curve over the terrain that the character should follow, or a target location for punching and kicking. The feedforward control network and the motion manifold are trained independently, allowing the user to easily switch between feedforward networks according to the desired interface, without re-training the motion manifold. Once motion is generated it can be edited by performing optimization in the space of the motion manifold. This allows for imposing kinematic constraints, or transforming the style of the motion, while ensuring the edited motion remains natural. As a result, the system can produce smooth, high quality motion sequences without any manual pre-processing of the training data.

Deep learning in medical imaging and radiation therapy.

Abstract: The goals of this review paper on deep learning (DL) in medical imaging and radiation therapy are to (a) summarize what has been achieved to date; (b) identify common and unique challenges, and strategies that researchers have taken to address these challenges; and (c) identify some of the promising avenues for the future both in terms of applications as well as technical innovations. We introduce the general principles of DL and convolutional neural networks, survey five major areas of application of DL in medical imaging and radiation therapy, identify common themes, discuss methods for dataset expansion, and conclude by summarizing lessons learned, remaining challenges, and future directions.