Ferdia Sherry

Ferdia Sherry

Auto-generated profile, based on publicly available information

Ferdia Sherry

University of Cambridge

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
17 August 2020 in IEEE Transactions on Medical Imaging
Verified
Journal Article
Closed access
FSFerdia Sherry
Ferdia Sherry
MBMartin Benning
Martin Benning
JRJuan Carlos De Los Reyes
Juan Carlos De Los Reyes
MGMartin J. Graves
Martin J. Graves
GMGeorg Maierhofer
Georg Maierhofer
MEMatthias J. Ehrhardt
Matthias J. Ehrhardt
The discovery of the theory of compressed sensing brought the realisation that many inverse problems can be solved even when measurements are "incomplete". This is particularly interesting in magnetic resonance imaging (MRI), where long acquisition times can limit its use. In this work, we consider the problem of learning a sparse sampling pattern that can be used to optimally balance acquisition time versus quality of the reconstructed image. We use a supervised learning approach, making the assumption that our training data is representative enough of new data acquisitions. We demonstrate that this is indeed the case, even if the training data consists of just 7 training pairs of measurements and ground-truth images; with a training set of brain images of size 192 by 192, for instance, one of the learned patterns samples only 35% of k-space, however results in reconstructions with mean SSIM 0.914 on a test set of similar images. The proposed framework is general enough to learn arbitrary sampling patterns, including common patterns such as Cartesian, spiral and radial sampling.

2 January 2023 in IEEE Signal Processing Magazine
Verified
Journal Article
Closed access
DCDongdong Chen
Dongdong Chen
MDMike Davies
Mike Davies
MEMatthias J. Ehrhardt
Matthias J. Ehrhardt
CSCarola-Bibiane Schonlieb
Carola-Bibiane Schonlieb
FSFerdia Sherry
Ferdia Sherry
JTJulian Tachella
Julian Tachella
From early image processing to modern computational imaging, successful models and algorithms have relied on a fundamental property of natural signals: symmetry . Here symmetry refers to the invariance property of signal sets to transformations, such as translation, rotation, or scaling. Symmetry can also be incorporated into deep neural networks (DNNs) in the form of equivariance, allowing for more data-efficient learning. While there have been important advances in the design of end-to-end equivariant networks for image classification in recent years, computational imaging introduces unique challenges for equivariant network solutions since we typically only observe the image through some noisy ill-conditioned forward operator that itself may not be equivariant. We review the emerging field of equivariant imaging (EI) and show how it can provide improved generalization and new imaging opportunities. Along the way, we show the interplay between the acquisition physics and group actions and links to iterative reconstruction, blind compressed sensing, and self-supervised learning.

27 May 2021 in European Journal of Applied Mathematics
Verified
Journal Article
Open access
PDF file
ECElena Cellodoni
Elena Cellodoni
MEM. J. Ehrhardt
M. J. Ehrhardt
CEC. Etmann
C. Etmann
RMR. I. Mclachlan
R. I. Mclachlan
BOB. Owren
B. Owren
FSF. Sherry
F. Sherry
Over the past few years, deep learning has risen to the foreground as a topic of massive interest, mainly as a result of successes obtained in solving large-scale image processing tasks. There are multiple challenging mathematical problems involved in applying deep learning: most deep learning methods require the solution of hard optimisation problems, and a good understanding of the trade-off between computational effort, amount of data and model complexity is required to successfully design a deep learning approach for a given problem.. A large amount of progress made in deep learning has been based on heuristic explorations, but there is a growing effort to mathematically understand the structure in existing deep learning methods and to systematically design new deep learning methods to preserve certain types of structure in deep learning. In this article, we review a number of these directions: some deep neural networks can be understood as discretisations of dynamical systems, neural networks can be designed to have desirable properties such as invertibility or group equivariance and new algorithmic frameworks based on conformal Hamiltonian systems and Riemannian manifolds to solve the optimisation problems have been proposed. We conclude our review of each of these topics by discussing some open problems that we consider to be interesting directions for future research.

26 July 2021 in Inverse Problems
Verified
Journal Article
Open access
PDF file
ECElena Celledoni
Elena Celledoni
MEMatthias J. Ehrhardt
Matthias J. Ehrhardt
CEChristian Etmann
Christian Etmann
BOBrynjulf Owren
Brynjulf Owren
CSCarola-Bibiane Schönlieb
Carola-Bibiane Schönlieb
FSFerdia Sherry
Ferdia Sherry
In recent years the use of convolutional layers to encode an inductive bias (translational equivariance) in neural networks has proven to be a very fruitful idea. The successes of this approach have motivated a line of research into incorporating other symmetries into deep learning methods, in the form of group equivariant convolutional neural networks. Much of this work has been focused on roto-translational symmetry of R d , but other examples are the scaling symmetry of R d and rotational symmetry of the sphere. In this work, we demonstrate that group equivariant convolutional operations can naturally be incorporated into learned reconstruction methods for inverse problems that are motivated by the variational regularisation approach. Indeed, if the regularisation functional is invariant under a group symmetry, the corresponding proximal operator will satisfy an equivariance property with respect to the same group symmetry. As a result of this observation, we design learned iterative methods in which the proximal operators are modelled as group equivariant convolutional neural networks. We use roto-translationally equivariant operations in the proposed methodology and apply it to the problems of low-dose computerised tomography reconstruction and subsampled magnetic resonance imaging reconstruction. The proposed methodology is demonstrated to improve the reconstruction quality of a learned reconstruction method with a little extra computational cost at training time but without any extra cost at test time.

3 June 2024 in Foundations of Computational Mathematics
Verified
Journal Article
Open access
PDF file
AHAndreas Hauptmann
Andreas Hauptmann
SMSubhadip Mukherjee
Subhadip Mukherjee
CSCarola-Bibiane Schönlieb
Carola-Bibiane Schönlieb
FSFerdia Sherry
Ferdia Sherry
Regularization is necessary when solving inverse problems to ensure the well-posedness of the solution map. Additionally, it is desired that the chosen regularization strategy is convergent in the sense that the solution map converges to a solution of the noise-free operator equation. This provides an important guarantee that stable solutions can be computed for all noise levels and that solutions satisfy the operator equation in the limit of vanishing noise. In recent years, reconstructions in inverse problems are increasingly approached from a data-driven perspective. Despite empirical success, the majority of data-driven approaches do not provide a convergent regularization strategy. One such popular example is given by iterative plug-and-play (PnP) denoising using off-the-shelf image denoisers. These usually provide only convergence of the PnP iterates to a fixed point, under suitable regularity assumptions on the denoiser, rather than convergence of the method as a regularization technique, thatis under vanishing noise and regularization strength. This paper serves two purposes: first, we provide an overview of the classical regularization theory in inverse problems and survey a few notable recent data-driven methods that are provably convergent regularization schemes. We then continue to discuss PnP algorithms and their established convergence guarantees. Subsequently, we consider PnP algorithms with learned linear denoisers and propose a novel spectral filtering technique of the denoiser to control the strength of regularization. Further, by relating the implicit regularization of the denoiser to an explicit regularization functional, we are the first to rigorously show that PnP with a learned linear denoiser leads to a convergent regularization scheme. The theoretical analysis is corroborated by numerical experiments for the classical inverse problem of tomographic image reconstruction.

15 April 2024 in Physica D: Nonlinear Phenomena
Verified
Journal Article
Open access
FSFerdia Sherry
Ferdia Sherry
ECElena Celledoni
Elena Celledoni
MEMatthias J. Ehrhardt
Matthias J. Ehrhardt
DMDavide Murari
Davide Murari
BOBrynjulf Owren
Brynjulf Owren
CSCarola-Bibiane Schönlieb
Carola-Bibiane Schönlieb