arXiv subjects

Fernando Maestu

Publications and source records attributed to Fernando Maestu.

3 recordsLinked to original sources

arXiv2020

A Graph Gaussian Embedding Method for Predicting Alzheimer's Disease Progression with MEG Brain Networks

Characterizing the subtle changes of functional brain networks associated with the pathological cascade of Alzheimer's disease (AD) is important for early diagnosis and prediction of disease progression prior to clinical symptoms. We developed a new deep learning method, termed multiple graph Gaussian embedding model (MG2G), which can learn highly informative network features by mapping high-dimensional resting-state brain networks into a low-dimensional latent space. These latent distribution-based embeddings enable a quantitative characterization of subtle and heterogeneous brain connectivity patterns at different regions and can be used as input to traditional classifiers for various downstream graph analytic tasks, such as AD early stage prediction, and statistical evaluation of between-group significant alterations across brain regions. We used MG2G to detect the intrinsic latent dimensionality of MEG brain networks, predict the progression of patients with mild cognitive impairment (MCI) to AD, and identify brain regions with network alterations related to MCI.

q-bio.NC↗

arXiv2020

FNS: an event-driven spiking neural network simulator based on the LIFL neuron model

Limitations in processing capabilities and memory of today's computers make spiking neuron-based (human) whole-brain simulations inevitably characterized by a compromise between bio-plausibility and computational cost. It translates into brain models composed of a reduced number of neurons and a simplified neuron's mathematical model, leading to the search for new simulation strategies. Taking advantage of the sparse character of brain-like computation, the event-driven technique could represent a way to carry out efficient simulation of large-scale Spiking Neural Networks (SNN). The recent Leaky Integrate-and-Fire with Latency (LIFL) spiking neuron model is event-driven compatible and exhibits some realistic neuronal features, opening new avenues for brain modelling. In this paper we introduce FNS, the first LIFL-based spiking neural network framework, which combines spiking/synaptic neural modelling with the event-driven approach, allowing us to define heterogeneous neuron modules and multi-scale connectivity with delayed connections and plastic synapses. In order to allow multi-thread implementations a novel parallelization strategy is also introduced. This paper presents mathematical models, software implementation and simulation routines on which FNS is based. Finally, a brain subnetwork is modeled on the basis of real brain structural data, and the resulting simulated activity is compared with associated brain functional (source-space MEG) data, demonstrating a good matching between the activity of the model and that of the experimetal data. This work aims to lay the groundwork for future event-driven based personalised brain models.

q-bio.NC↗

arXiv2018

A neuro-inspired system for online learning and recognition of parallel spike trains, based on spike latency and heterosynaptic STDP

Humans perform remarkably well in many cognitive tasks including pattern recognition. However, the neuronal mechanisms underlying this process are not well understood. Nevertheless, artificial neural networks, inspired in brain circuits, have been designed and used to tackle spatio-temporal pattern recognition tasks. In this paper we present a multineuronal spike pattern detection structure able to autonomously implement online learning and recognition of parallel spike sequences (i.e., sequences of pulses belonging to different neurons/neural ensembles). The operating principle of this structure is based on two spiking/synaptic neurocomputational characteristics: spike latency, that enables neurons to fire spikes with a certain delay and heterosynaptic plasticity, that allows the own regulation of synaptic weights. From the perspective of the information representation, the structure allows mapping a spatio-temporal stimulus into a multidimensional, temporal, feature space. In this space, the parameter coordinate and the time at which a neuron fires represent one specific feature. In this sense, each feature can be considered to span a single temporal axis. We applied our proposed scheme to experimental data obtained from a motor inhibitory cognitive task. The test exhibits good classification performance, indicating the adequateness of our approach. In addition to its effectiveness, its simplicity and low computational cost suggest a large scale implementation for real time recognition applications in several areas, such as brain computer interface, personal biometrics authentication or early detection of diseases.

q-bio.NC↗