Filters








25 Hits in 1.2 sec

Algorithmic Amplification of Politics on Twitter [article]

Ferenc Huszár, Sofia Ira Ktena, Conor O'Brien, Luca Belli, Andrew Schlaikjer, Moritz Hardt
2021 arXiv   pre-print
Supplementary Information for Algorithmic Amplification of Politics on Twitter Ferenc Huszár, Sofia Ira Ktena, Conor O'Brien, Luca Belli, Andrew Schlaikjer and Moritz Hardt identified the appropriate  ... 
arXiv:2110.11010v1 fatcat:wweygf6acjb7jfnhgui3hbssay

Comparison of Brain Networks with Unknown Correspondences [article]

Sofia Ira Ktena, Sarah Parisot, Jonathan Passerat-Palmbach, Daniel Rueckert
2016 arXiv   pre-print
Graph theory has drawn a lot of attention in the field of Neuroscience during the last decade, mainly due to the abundance of tools that it provides to explore the interactions of elements in a complex network like the brain. The local and global organization of a brain network can shed light on mechanisms of complex cognitive functions, while disruptions within the network can be linked to neurodevelopmental disorders. In this effort, the construction of a representative brain network for each
more » ... individual is critical for further analysis. Additionally, graph comparison is an essential step for inference and classification analyses on brain graphs. In this work we explore a method based on graph edit distance for evaluating graph similarity, when correspondences between network elements are unknown due to different underlying subdivisions of the brain. We test this method on 30 unrelated subjects as well as 40 twin pairs and show that this method can accurately reflect the higher similarity between two related networks compared to unrelated ones, while identifying node correspondences.
arXiv:1611.04783v1 fatcat:rniydy7bcfakrfxweqyy2dsqpu

Deep Bayesian Bandits: Exploring in Online Personalized Recommendations [article]

Dalin Guo, Sofia Ira Ktena, Ferenc Huszar, Pranay Kumar Myana, Wenzhe Shi, Alykhan Tejani
2020 arXiv   pre-print
Recommender systems trained in a continuous learning fashion are plagued by the feedback loop problem, also known as algorithmic bias. This causes a newly trained model to act greedily and favor items that have already been engaged by users. This behavior is particularly harmful in personalised ads recommendations, as it can also cause new campaigns to remain unexplored. Exploration aims to address this limitation by providing new information about the environment, which encompasses user
more » ... nce, and can lead to higher long-term reward. In this work, we formulate a display advertising recommender as a contextual bandit and implement exploration techniques that require sampling from the posterior distribution of click-through-rates in a computationally tractable manner. Traditional large-scale deep learning models do not provide uncertainty estimates by default. We approximate these uncertainty measurements of the predictions by employing a bootstrapped model with multiple heads and dropout units. We benchmark a number of different models in an offline simulation environment using a publicly available dataset of user-ads engagements. We test our proposed deep Bayesian bandits algorithm in the offline simulation and online AB setting with large-scale production traffic, where we demonstrate a positive gain of our exploration model.
arXiv:2008.00727v1 fatcat:eu5rpcg6abhmxlmg4ewnbi7zga

Graph Saliency Maps through Spectral Convolutional Networks: Application to Sex Classification with Brain Connectivity [article]

Salim Arslan, Sofia Ira Ktena, Ben Glocker, Daniel Rueckert
2018 arXiv   pre-print
Graph convolutional networks (GCNs) allow to apply traditional convolution operations in non-Euclidean domains, where data are commonly modelled as irregular graphs. Medical imaging and, in particular, neuroscience studies often rely on such graph representations, with brain connectivity networks being a characteristic example, while ultimately seeking the locus of phenotypic or disease-related differences in the brain. These regions of interest (ROIs) are, then, considered to be closely
more » ... ted with function and/or behaviour. Driven by this, we explore GCNs for the task of ROI identification and propose a visual attribution method based on class activation mapping. By undertaking a sex classification task as proof of concept, we show that this method can be used to identify salient nodes (brain regions) without prior node labels. Based on experiments conducted on neuroimaging data of more than 5000 participants from UK Biobank, we demonstrate the robustness of the proposed method in highlighting reproducible regions across individuals. We further evaluate the neurobiological relevance of the identified regions based on evidence from large-scale UK Biobank studies.
arXiv:1806.01764v1 fatcat:orqgczzaxzg75b4tup2dca5epq

Metric learning with spectral graph convolutions on brain connectivity networks

Sofia Ira Ktena, Sarah Parisot, Enzo Ferrante, Martin Rajchl, Matthew Lee, Ben Glocker, Daniel Rueckert
2018 NeuroImage  
., 2010; Ktena et al., 2017a) , but are limited by the 65 a priori definition of the edit costs.  ...  We extend the preliminary work on metric learning for irregular 555 graphs (Ktena et al., 2017b) and perform a more thorough evaluation of the proposed method.  ... 
doi:10.1016/j.neuroimage.2017.12.052 pmid:29278772 fatcat:ihthm266gbclpcm3eqq3tu2zmm

DLTK: State of the Art Reference Implementations for Deep Learning on Medical Images [article]

Nick Pawlowski, Sofia Ira Ktena, Matthew C.H. Lee, Bernhard Kainz, Daniel Rueckert, Ben Glocker, Martin Rajchl
2017 arXiv   pre-print
We present DLTK, a toolkit providing baseline implementations for efficient experimentation with deep learning methods on biomedical images. It builds on top of TensorFlow and its high modularity and easy-to-use examples allow for a low-threshold access to state-of-the-art implementations for typical medical imaging problems. A comparison of DLTK's reference implementations of popular network architectures for image segmentation demonstrates new top performance on the publicly available
more » ... e data "Multi-Atlas Labeling Beyond the Cranial Vault". The average test Dice similarity coefficient of 81.5 exceeds the previously best performing CNN (75.7) and the accuracy of the challenge winning method (79.0).
arXiv:1711.06853v1 fatcat:io3culrpajg7zahamdq2cvxklm

Spectral Graph Convolutions for Population-Based Disease Prediction [chapter]

Sarah Parisot, Sofia Ira Ktena, Enzo Ferrante, Matthew Lee, Ricardo Guerrerro Moreno, Ben Glocker, Daniel Rueckert
2017 Lecture Notes in Computer Science  
Exploiting the wealth of imaging and non-imaging information for disease prediction tasks requires models capable of representing, at the same time, individual features as well as data associations between subjects from potentially large populations. Graphs provide a natural framework for such tasks, yet previous graph-based approaches focus on pairwise similarities without modelling the subjects' individual characteristics and features. On the other hand, relying solely on subjectspecific
more » ... ng feature vectors fails to model the interaction and similarity between subjects, which can reduce performance. In this paper, we introduce the novel concept of Graph Convolutional Networks (GCN) for brain analysis in populations, combining imaging and non-imaging data. We represent populations as a sparse graph where its vertices are associated with image-based feature vectors and the edges encode phenotypic information. This structure was used to train a GCN model on partially labelled graphs, aiming to infer the classes of unlabelled nodes from the node features and pairwise associations between subjects. We demonstrate the potential of the method on the challenging ADNI and ABIDE databases, as a proof of concept of the benefit from integrating contextual information in classification tasks. This has a clear impact on the quality of the predictions, leading to 69.5% accuracy for ABIDE (outperforming the current state of the art of 66.8%) and 77% for ADNI for prediction of MCI conversion, significantly outperforming standard linear classifiers where only individual features are considered.
doi:10.1007/978-3-319-66179-7_21 fatcat:nqad7xynczfabmba7dxrgk2thu

Human brain mapping: A systematic comparison of parcellation methods for the human cerebral cortex

Salim Arslan, Sofia Ira Ktena, Antonios Makropoulos, Emma C. Robinson, Daniel Rueckert, Sarah Parisot
2018 NeuroImage  
The macro-connectome elucidates the pathways through which brain regions are structurally connected or functionally coupled to perform a specific cognitive task. It embodies the notion of representing and understanding all connections within the brain as a network, while the subdivision of the brain into interacting functional units is inherent in its architecture. As a result, the definition of network nodes is one of the most critical steps in connectivity network analysis. Although brain
more » ... ses obtained from cytoarchitecture or anatomy have long been used for this task, connectivity-driven methods have arisen only recently, aiming to delineate more homogeneous and functionally coherent regions. This study provides a systematic comparison between anatomical, connectivitydriven and random parcellation methods proposed in the thriving field of brain parcellation. Using functional MRI data from the Human Connectome Project and a plethora of quantitative evaluation techniques investigated in the literature, we evaluate 10 subject-level and 24 groupwise parcellation methods at different resolutions. We assess the accuracy of parcellations from four different aspects: (1) reproducibility across different acquisitions and groups, (2) fidelity to the underlying connectivity data, (3) agreement with fMRI task activation, myelin maps, and cytoarchitectural areas, and (4) network analysis. This extensive evaluation of different parcellations generated at the subject and group the underlying data, such methods can potentially provide highly homogeneous and functionally coherent parcels and separate regions with different patterns of connectivity more accurately. With this idea in mind, several connectivitydriven parcellation methods have been proposed, usually in association with clustering techniques (Thirion et al., 2014) . These methods are based on hier-50 archical clustering (Mumford et al., 2010; Bellec et al., 2010; Moreno-Dominguez et al., 2014) , k -means (and its fuzzy counterpart) (Tomassini et al., 2007; Mezer et al., 2009; Golland et al., 2008) , Gaus-155 is carried out separately for each of the 86 different functional contrasts, over 7 different tasks, including the motor protocol, the relational protocol, the social protocol, the language protocol, the emotion protocol, the gambling protocol, and the working memory protocol ( Barch et al., 2013) . The analysis is performed across sessions (single subject activation maps) and then across subjects 160 (groupwise activation map). We compute the group average myelin maps by averaging all subjects' myelin maps, while the average Brodmann map is obtained implementations are available (Honnorat et al., 2015; Parisot et al., 2016b) .
doi:10.1016/j.neuroimage.2017.04.014 pmid:28412442 fatcat:yrf2qf42snfkjlec6twcsba3ja

Brain connectivity measures improve modeling of functional outcome after acute ischemic stroke [article]

Sofia Ira Ktena, Markus D. Schirmer, Mark R. Etherton, Anne-Katrin Giese, Carissa Tuozzo, Brittany B Mills, Daniel Rueckert, Ona Wu, Natalia S. Rost
2019 bioRxiv   pre-print
Sofia Ira Ktena was supported by the EPSRC Centre for Doctoral Training in High Performance Embedded and Distributed Systems (HiPEDS, Grant Reference EP/L016796/1) and an EMBO short-term fellowship (Reference  ... 
doi:10.1101/590497 fatcat:e2omxprafzfqpdsyg2mi33tv5i

Exploring heritability of functional brain networks with inexact graph matching

Sofia Ira Ktena, Salim Arslan, Sarah Parisot, Daniel Rueckert
2017 2017 IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017)  
Data-driven brain parcellations aim to provide a more accurate representation of an individual's functional connectivity, since they are able to capture individual variability that arises due to development or disease. This renders comparisons between the emerging brain connectivity networks more challenging, since correspondences between their elements are not preserved. Unveiling these correspondences is of major importance to keep track of local functional connectivity changes. We propose a
more » ... ovel method based on graph edit distance for the comparison of brain graphs directly in their domain, that can accurately reflect similarities between individual networks while providing the network element correspondences. This method is validated on a dataset of 116 twin subjects provided by the Human Connectome Project.
doi:10.1109/isbi.2017.7950536 dblp:conf/isbi/KtenaAPR17 fatcat:xsujg2bg3bat5eawtkcyfc5nxq

Rich-Club organization: an important determinant of functional outcome after acute ischemic stroke [article]

Markus D Schirmer, Sofia Ira Ktena, Marco J Nardin, Kathleen L Donahue, Anne-Katrin Giese, Mark R Etherton, Ona Wu, Natalia S Rost
2019 bioRxiv   pre-print
Objective: To determine whether the rich-club organization, essential for information transport in the human connectome, is an important biomarker of functional outcome after acute ischemic stroke (AIS). Methods: Consecutive AIS patients (N=344) with acute brain magnetic resonance imaging (MRI) (<48 hours) were eligible for this study. Each patient underwent a clinical MRI protocol, which included diffusion weighted imaging (DWI). All DWIs were registered to a template on which rich-club
more » ... have been defined. Using manual outlines of stroke lesions, we automatically counted the number of affected rich-club regions and assessed its effect on the National Institute of Health Stroke Scale (NIHSS) and modified Rankin Scale (mRS; obtained at 90 days post-stroke) scores through ordinal regression. Results: Of 344 patients (median age 65, inter-quartile range 54-76 years) with a median DWI lesion volume (DWIv) of 3cc, 64% were male. We established that an increase in number of rich-club regions affected by a stroke increases the odds of poor stroke outcome, measured by NIHSS (OR: 1.77, 95%CI 1.41-2.21) and mRS (OR: 1.38, 95%CI 1.11-1.73). Additionally, we demonstrated that the OR exceeds traditional markers, such as DWIv (OR(NIHSS) 1.08, 95%CI 1.06-1.11; OR(mRS) 1.05, 95%CI 1.03-1.07) and age (OR(NIHSS) 1.03, 95%CI 1.01-1.05; OR(mRS) 1.05, 95%CI 1.03-1.07). Conclusion: In this proof-of-concept study, the number of rich-club nodes affected by a stroke lesion presents a translational biomarker of stroke outcome, which can be readily assessed using standard clinical AIS imaging protocols and considered in functional outcome prediction models beyond traditional factors.
doi:10.1101/545897 fatcat:4aiapl2jrzb7tfzijqzmgw7fry

Model Size Reduction Using Frequency Based Double Hashing for Recommender Systems [article]

Caojin Zhang, Yicun Liu, Yuanpu Xie, Sofia Ira Ktena, Alykhan Tejani, Akshay Gupta, Pranay Kumar Myana, Deepak Dilipkumar, Suvadip Paul, Ikuhiro Ihara, Prasang Upadhyaya, Ferenc Huszar (+1 others)
2020 arXiv   pre-print
[Ktena et al., 2019] .  ... 
arXiv:2007.14523v1 fatcat:oh7xmkcu5jdgdk6uacpo2aiyle

Distance Metric Learning Using Graph Convolutional Networks: Application to Functional Brain Networks [chapter]

Sofia Ira Ktena, Sarah Parisot, Enzo Ferrante, Martin Rajchl, Matthew Lee, Ben Glocker, Daniel Rueckert
2017 Lecture Notes in Computer Science  
Evaluating similarity between graphs is of major importance in several computer vision and pattern recognition problems, where graph representations are often used to model objects or interactions between elements. The choice of a distance or similarity metric is, however, not trivial and can be highly dependent on the application at hand. In this work, we propose a novel metric learning method to evaluate distance between graphs that leverages the power of convolutional neural networks, while
more » ... xploiting concepts from spectral graph theory to allow these operations on irregular graphs. We demonstrate the potential of our method in the field of connectomics, where neuronal pathways or functional connections between brain regions are commonly modelled as graphs. In this problem, the definition of an appropriate graph similarity function is critical to unveil patterns of disruptions associated with certain brain disorders. Experimental results on the ABIDE dataset show that our method can learn a graph similarity metric tailored for a clinical application, improving the performance of a simple k-nn classifier by 11.9% compared to a traditional distance metric.
doi:10.1007/978-3-319-66182-7_54 fatcat:ydvxdnwacjf5vdoa4dbnfavp7e

Distance Metric Learning using Graph Convolutional Networks: Application to Functional Brain Networks [article]

Sofia Ira Ktena, Sarah Parisot, Enzo Ferrante, Martin Rajchl, Matthew Lee, Ben Glocker, Daniel Rueckert
2017 arXiv   pre-print
Evaluating similarity between graphs is of major importance in several computer vision and pattern recognition problems, where graph representations are often used to model objects or interactions between elements. The choice of a distance or similarity metric is, however, not trivial and can be highly dependent on the application at hand. In this work, we propose a novel metric learning method to evaluate distance between graphs that leverages the power of convolutional neural networks, while
more » ... xploiting concepts from spectral graph theory to allow these operations on irregular graphs. We demonstrate the potential of our method in the field of connectomics, where neuronal pathways or functional connections between brain regions are commonly modelled as graphs. In this problem, the definition of an appropriate graph similarity function is critical to unveil patterns of disruptions associated with certain brain disorders. Experimental results on the ABIDE dataset show that our method can learn a graph similarity metric tailored for a clinical application, improving the performance of a simple k-nn classifier by 11.9% compared to a traditional distance metric.
arXiv:1703.02161v2 fatcat:uaeuq6g4ircydiwipqkjgdszeq

Spectral Graph Convolutions for Population-based Disease Prediction [article]

Sarah Parisot, Sofia Ira Ktena, Enzo Ferrante, Matthew Lee, Ricardo Guerrerro Moreno, Ben Glocker, Daniel Rueckert
2017 arXiv   pre-print
Exploiting the wealth of imaging and non-imaging information for disease prediction tasks requires models capable of representing, at the same time, individual features as well as data associations between subjects from potentially large populations. Graphs provide a natural framework for such tasks, yet previous graph-based approaches focus on pairwise similarities without modelling the subjects' individual characteristics and features. On the other hand, relying solely on subject-specific
more » ... ing feature vectors fails to model the interaction and similarity between subjects, which can reduce performance. In this paper, we introduce the novel concept of Graph Convolutional Networks (GCN) for brain analysis in populations, combining imaging and non-imaging data. We represent populations as a sparse graph where its vertices are associated with image-based feature vectors and the edges encode phenotypic information. This structure was used to train a GCN model on partially labelled graphs, aiming to infer the classes of unlabelled nodes from the node features and pairwise associations between subjects. We demonstrate the potential of the method on the challenging ADNI and ABIDE databases, as a proof of concept of the benefit from integrating contextual information in classification tasks. This has a clear impact on the quality of the predictions, leading to 69.5% accuracy for ABIDE (outperforming the current state of the art of 66.8%) and 77% for ADNI for prediction of MCI conversion, significantly outperforming standard linear classifiers where only individual features are considered.
arXiv:1703.03020v3 fatcat:khiaorircfexhp4x5tf4icqj3i
« Previous Showing results 1 — 15 out of 25 results