Time filter

Source Type

Alvarez M.A.,Technological University of Pereira | Luengo D.,Technical University of Madrid | Lawrence N.D.,University of Sheffield | Lawrence N.D.,Sheffield Institute for Translational Neuroscience
IEEE Transactions on Pattern Analysis and Machine Intelligence | Year: 2013

Purely data-driven approaches for machine learning present difficulties when data are scarce relative to the complexity of the model or when the model is forced to extrapolate. On the other hand, purely mechanistic approaches need to identify and specify all the interactions in the problem at hand (which may not be feasible) and still leave the issue of how to parameterize the system. In this paper, we present a hybrid approach using Gaussian processes and differential equations to combine data-driven modeling with a physical model of the system. We show how different, physically inspired, kernel functions can be developed through sensible, simple, mechanistic assumptions about the underlying system. The versatility of our approach is illustrated with three case studies from motion capture, computational biology, and geostatistics. © 1979-2012 IEEE.

Crunelli V.,University of Cardiff | Cope D.W.,University of Cardiff | Terry J.R.,University of Sheffield | Terry J.R.,Sheffield Institute for Translational Neuroscience
Epilepsy Research | Year: 2011

Absence seizures appear to be initiated in a putative cortical 'initiation site' by the expression of medium-amplitude 5-9Hz oscillations, which may in part be due to a decreased phasic GABA A receptor function. These oscillations rapidly spread to other cortical areas and to the thalamus, leading to fully developed generalized spike and wave discharges. In thalamocortical neurons of genetic models, phasic GABA A inhibition is either unchanged or increased, whereas tonic GABA A inhibition is increased both in genetic and pharmacological models. This enhanced tonic inhibition is required for absence seizure generation, and in genetic models it results from a malfunction in the astrocytic GABA transporter GAT-1. Contradictory results from inbred and transgenic animals still do not allow us to draw firm conclusions on changes in phasic GABA A inhibition in the GABAergic neurons of the nucleus reticularis thalami. Mathematical modelling may enhance our understanding of these competing hypotheses, by permitting investigations of their mechanistic aspects, hence enabling a greater understanding of the processes underlying seizure generation and evolution. © 2011 Elsevier B.V..

Alvarez M.A.,University of Manchester | Peters J.,Max Planck Institute for Biological Cybernetics | Scholkopf B.,Max Planck Institute for Biological Cybernetics | Lawrence N.D.,University of Sheffield | Lawrence N.D.,Sheffield Institute for Translational Neuroscience
Advances in Neural Information Processing Systems 23: 24th Annual Conference on Neural Information Processing Systems 2010, NIPS 2010 | Year: 2010

Latent force models encode the interaction between multiple related dynamical systems in the form of a kernel or covariance function. Each variable to be modeled is represented as the output of a differential equation and each differential equation is driven by a weighted sum of latent functions with uncertainty given by a Gaussian process prior. In this paper we consider employing the latent force model framework for the problem of determining robot motor primitives. To deal with discontinuities in the dynamical systems or the latent driving force we introduce an extension of the basic latent force model, that switches between different latent functions and potentially different dynamical systems. This creates a versatile representation for robot movements that can capture discrete changes and non-linearities in the dynamics. We give illustrative examples on both synthetic data and for striking movements recorded using a BarrettWAM robot as haptic input device. Our inspiration is robot motor primitives, but we expect our model to have wide application for dynamical systems including models for human motion capture data and systems biology.

Zaccai J.,University of Cambridge | Brayne C.,University of Cambridge | Matthews F.E.,Institute of Public Health | Ince P.G.,Sheffield Institute for Translational Neuroscience
Alzheimer's Research and Therapy | Year: 2015

Introduction: Studies with strong selection biases propose that alpha-synucleinopathy (AS) spreads upwards and downwards in the neuraxis from the medulla, that amygdala-dominant AS is strongly associated with Alzheimer's disease (AD), and that a more severe involvement of the cerebral cortex is correlated with increasing risk of dementia. This study examines the association of AS patterns and observed neuropsychological symptoms in brains of a population-representative donor cohort. Methods: Brains donated in 2 out of 6 cognitive function and ageing study cohorts (Cambridgeshire and Nottingham) were examined. Over 80% were >80 years old at death. The respondents were evaluated prospectively in life for cognitive decline and dementia. Immunocytochemistry for tau and alpha-synuclein (using LB509 by Zymed Laboratories) was carried out in 208 brains to establish Braak stage and the pattern and severity of AS following the dementia with Lewy bodies (DLB) consensus recommendations. Dementia, specific neuropsychological measures as measured using the Cambridge cognitive examination, the presence of hallucinations and Parkinson's disease were investigated. Results: Four patterns of AS were observed: no AS pathology (n = 92), AS pathology following the DLB consensus guidelines (n = 33, of which five were 'neocortical'), amygdala-predominant AS (n = 18), and other AS patterns (n = 33). Each group was subdivided according to high/low neurofibrillary tangles (NFT) Braak stage. Results showed no association between dementia and these patterns of AS, adjusting for the presence of NFT or not. The risk of visual hallucinations shows a weak association with AS in the substantia nigra (odds ratio (OR) = 3.2; 95% confidence interval (CI) 0.5 to 15.5; P = 0.09) and amygdala (OR = 3.0; 95% CI 0.7 to 12.3; P = 0.07). The analysis is similar for auditory hallucinations in subcortical regions. Conclusions: Among the whole population of older people, AS does not increase the risks for dementia, irrespective of Braak stage of NFT pathology. There was no evidence that the pattern of AS pathology in cortical areas was relevant to the risk of hallucination. In general, the hypothesis that AS as measured using these methods per se is a key determinant of cognitive clinical phenotypes is not supported. © 2015 Zaccai et al.; licensee BioMed Central.

Cooper-Knock J.,Academic Unit of Neurology | Cooper-Knock J.,Sheffield Institute for Translational Neuroscience | Ahmedzai S.H.,University of Sheffield | Shaw P.,Academic Unit of Neurology
Amyotrophic Lateral Sclerosis | Year: 2011

Sialorrhoea is a recognized complication of bulbar amyotrophic lateral sclerosis (ALS) that leads to an increased risk of potentially harmful aspiration and often prevents patients from tolerating non-invasive ventilation (NIV). A case of treatment-resistant sialorrhoea in bulbar ALS is described where subcutaneous glycopyrrolate was effective without significant side-effects. The patient went on to markedly increase the length of time she could tolerate NIV each night. © 2011 Informa Healthcare.

Discover hidden collaborations