Author name cluster

Fernando Maestú

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

16 papers

1 author row

JBHI Journal 2025 Journal Article

Fully Hyperbolic Neural Networks: A Novel Approach to Studying Aging Trajectories

Hugo Ramirez
Davide Tabarelli
Arianna Brancaccio
Paolo Belardinelli
Elisabeth B. Marsh
Michael Funke
John C. Mosher
Fernando Maestú

Characterizing age-related alterations in brain networks is crucial for understanding aging trajectories and identifying deviations indicative of neurodegenerative disorders, such as Alzheimer's disease. In this study, we developed a Fully Hyperbolic Neural Network (FHNN) to embed functional brain connectivity graphs derived from magnetoencephalography (MEG) data into low dimensions on a Lorentz model of hyperbolic space. Using this model, we computed hyperbolic embeddings of the MEG brain networks of 587 individuals from the Cambridge Centre for Ageing and Neuroscience (Cam-CAN) dataset. Notably, we leveraged a unique metric—the radius of the node embeddings—which effectively captures the hierarchical organization of the brain, to characterize subtle hierarchical organizational changes in various brain subnetworks attributed to the aging process. Our findings revealed that a considerable number of subnetworks exhibited a reduction in hierarchy during aging, with some showing gradual changes and others undergoing rapid transformations in the elderly. Moreover, we demonstrated that hyperbolic features outperform traditional graph-theoretic measures in capturing age-related information in brain networks. Overall, our study represents the first evaluation of hyperbolic embeddings in MEG brain networks for studying aging trajectories, shedding light on critical regions undergoing significant age-related alterations in the large cohort of the Cam-CAN dataset.

JBHI Journal 2024 Journal Article

Hyperbolic Graph Embedding of MEG Brain Networks to Study Brain Alterations in Individuals With Subjective Cognitive Decline

Cole Baker
Isabel Suárez-Méndez
Grace Smith
Elisabeth B. Marsh
Michael Funke
John C. Mosher
Fernando Maestú
Mengjia Xu

An expansive area of research focuses on discerning patterns of alterations in functional brain networks from the early stages of Alzheimer's disease, even at the subjective cognitive decline (SCD) stage. Here, we developed a novel hyperbolic MEG brain network embedding framework for transforming high-dimensional complex MEG brain networks into lower-dimensional hyperbolic representations. Using this model, we computed hyperbolic embeddings of the MEG brain networks of two distinct participant groups: individuals with SCD and healthy controls. We demonstrated that these embeddings preserve both local and global geometric information, presenting reduced distortion compared to rival models, even when brain networks are mapped into low-dimensional spaces. In addition, our findings showed that the hyperbolic embeddings encompass unique SCD-related information that improves the discriminatory power above and beyond that of connectivity features alone. Notably, we introduced a unique metric–the radius of the node embeddings–which effectively proxies the hierarchical organization of the brain. Using this metric, we identified subtle hierarchy organizational differences between the two participant groups, suggesting increased hierarchy in the dorsal attention, frontoparietal, and ventral attention subnetworks among the SCD group. Last, we assessed the correlation between these hierarchical variations and cognitive assessment scores, revealing associations with diminished performance across multiple cognitive evaluations in the SCD group. Overall, this study presents the first evaluation of hyperbolic embeddings of MEG brain networks, offering novel insights into brain organization, cognitive decline, and potential diagnostic avenues of Alzheimer's disease.

YNICL Journal 2022 Journal Article

Understanding brain function in vascular cognitive impairment and dementia with EEG and MEG: A systematic review

Lucía Torres-Simón
Sandra Doval
Alberto Nebreda
Sophia J. Llinas
Elisabeth B. Marsh
Fernando Maestú

Vascular Cognitive Impairment (VCI) is the second most prevalent dementia after Alzheimer's Disease (AD), and cerebrovascular disease (CBVD) is a major comorbid contributor to the progression of most neurodegenerative diseases. Early differentiation of cognitive impairment is critical given both the high prevalence of CBVD, and that its risk factors are modifiable. The ability for electroencephalogram (EEG) and magnetoencephalogram (MEG) to detect changes in brain functioning for other dementias suggests that they may also be promising biomarkers for early VCI. The present systematic review aims to summarize the literature regarding electrophysiological patterns of mild and major VCI. Despite considerable heterogeneity in clinical definition and electrophysiological methodology, common patterns exist when comparing patients with VCI to healthy controls (HC) and patients with AD, though there is a low specificity when comparing between VCI subgroups. Similar to other dementias, slowed frequency patterns and disrupted inter- and intra-hemispheric connectivity are repeatedly reported for VCI patients, as well as longer latencies and smaller amplitudes in evoked responses. Further study is needed to fully establish MEG and EEG as clinically useful biomarkers, including a clear definition of VCI and standardized methodology, allowing for comparison across groups and consolidation of multicenter efforts.

YNICL Journal 2019 Journal Article

Aberrant MEG multi-frequency phase temporal synchronization predicts conversion from mild cognitive impairment-to-Alzheimer's disease

Sandra Pusil
Stavros I. Dimitriadis
María Eugenia López
Ernesto Pereda
Fernando Maestú

Many neuroimaging studies focus on a frequency-specific or a multi-frequency network analysis showing that functional brain networks are disrupted in patients with Alzheimer's disease (AD). Although those studies enriched our knowledge of the impact of AD in brain's functionality, our goal is to test the effectiveness of combining neuroimaging with network neuroscience to predict with high accuracy subjects with mild cognitive impairment (MCI) that will convert to AD. In this study, eyes-closed resting-state magnetoencephalography (MEG) recordings from 27 stable MCI (sMCI) and 27 progressive MCI (pMCI) from two scan sessions (baseline and follow-up after approximately 3 years) were projected via beamforming onto an atlas-based set of regions of interest (ROIs). Dynamic functional connectivity networks were constructed independently for the five classical frequency bands while a multivariate phase-based coupling metric was adopted. Thus, computing the distance between the fluctuation of functional strength of every pair of ROIs between the two conditions with dynamic time wrapping (DTW), a large set of features was extracted. A machine learning algorithm revealed 49 DTW-based features in the five frequency bands that can distinguish the sMCI from pMCI with absolute accuracy (100%). Further analysis of the selected links revealed that most of the connected ROIs were part of the default mode network (DMN), the cingulo-opercular (CO), the fronto-parietal and the sensorimotor network. Overall, our dynamic network multi-frequency analysis approach provides an effective framework of constructing a sensitive MEG-based connectome biomarker for the prediction of conversion from MCI to Alzheimer's disease.

YNIMG Journal 2019 Journal Article

Dynamic low frequency EEG phase synchronization patterns during proactive control of task switching

María Eugenia López
Sandra Pusil
Ernesto Pereda
Fernando Maestú
Francisco Barceló

Cognitive flexibility is critical for humans living in complex societies with ever-growing multitasking demands. Yet the low-frequency neural dynamics of distinct task-specific and domain-general mechanisms sub-serving mental flexibility are still ill-defined. Here we estimated phase electroencephalogram synchronization by using inter-trial phase coherence (ITPC) at the source space while twenty six young participants were intermittently cued to switch or repeat their perceptual categorization rule of Gabor gratings varying in color and thickness (switch task). Therefore, the aim of this study was to examine whether a proactive control is associated with connectivity only in the frontoparietal theta network, or also involves distinct neural connectivity within the delta band, as distinct neural signatures while preparing to switch or repeat a task set, respectively. To this end, we focused the analysis on late-latencies (from 500 to 800 msec post-cue onset), since they are known to be associated with top-down cognitive control processes. We confirmed that proactive control during a task switch was associated with frontoparietal theta connectivity. But importantly, we also found a distinct role of delta band oscillatory synchronization in proactive control, engaging more posterior frontotemporal regions as opposed to frontoparietal theta connectivity. Additionally, we built a regression model by using the ITPC results in delta and theta bands as predictors, and the behavioral accuracy in the switch task as the criterion, obtaining significant results for both frequency bands. All these findings support the existence of distinct proactive cognitive control processes related to functionally distinct though highly complementary theta and delta frontoparietal and temporoparietal oscillatory networks at late-latency temporal scales.

YNICL Journal 2015 Journal Article

A multicenter study of the early detection of synaptic dysfunction in Mild Cognitive Impairment using Magnetoencephalography-derived functional connectivity

Fernando Maestú
Jose-Maria Peña
Pilar Garcés
Santiago González
Ricardo Bajo
Anto Bagic
Pablo Cuesta
Michael Funke

Synaptic disruption is an early pathological sign of the neurodegeneration of Dementia of the Alzheimer's type (DAT). The changes in network synchronization are evident in patients with Mild Cognitive Impairment (MCI) at the group level, but there are very few Magnetoencephalography (MEG) studies regarding discrimination at the individual level. In an international multicenter study, we used MEG and functional connectivity metrics to discriminate MCI from normal aging at the individual person level. A labeled sample of features (links) that distinguished MCI patients from controls in a training dataset was used to classify MCI subjects in two testing datasets from four other MEG centers. We identified a pattern of neuronal hypersynchronization in MCI, in which the features that best discriminated MCI were fronto-parietal and interhemispheric links. The hypersynchronization pattern found in the MCI patients was stable across the five different centers, and may be considered an early sign of synaptic disruption and a possible preclinical biomarker for MCI/DAT.

YNICL Journal 2015 Journal Article

What graph theory actually tells us about resting state interictal MEG epileptic activity

Guiomar Niso
Sira Carrasco
María Gudín
Fernando Maestú
Francisco del-Pozo
Ernesto Pereda

Graph theory provides a useful framework to study functional brain networks from neuroimaging data. In epilepsy research, recent findings suggest that it offers unique insight into the fingerprints of this pathology on brain dynamics. Most studies hitherto have focused on seizure activity during focal epilepsy, but less is known about functional epileptic brain networks during interictal activity in frontal focal and generalized epilepsy. Besides, it is not clear yet which measures are most suitable to characterize these networks. To address these issues, we recorded magnetoencephalographic (MEG) data using two orthogonal planar gradiometers from 45 subjects from three groups (15 healthy controls (7 males, 24 ± 6 years), 15 frontal focal (8 male, 32 ± 16 years) and 15 generalized epileptic (6 male, 27 ± 7 years) patients) during interictal resting state with closed eyes. Then, we estimated the total and relative spectral power of the largest principal component of the gradiometers, and the degree of phase synchronization between each sensor site in the frequency range [0. 5–40 Hz]. We further calculated a comprehensive battery of 15 graph-theoretic measures and used the affinity propagation clustering algorithm to elucidate the minimum set of them that fully describe these functional brain networks. The results show that differences in spectral power between the control and the other two groups have a distinctive pattern: generalized epilepsy presents higher total power for all frequencies except the alpha band over a widespread set of sensors; frontal focal epilepsy shows higher relative power in the beta band bilaterally in the fronto-central sensors. Moreover, all network indices can be clustered into three groups, whose exemplars are the global network efficiency, the eccentricity and the synchronizability. Again, the patterns of differences were clear: the brain network of the generalized epilepsy patients presented greater efficiency and lower eccentricity than the control subjects for the high frequency bands, without a clear topography. Besides, the frontal focal epileptic patients showed only reduced eccentricity for the theta band over fronto-temporal and central sensors. These outcomes indicate that functional epileptic brain networks are different to those of healthy subjects during interictal stage at rest, with a unique pattern of dissimilarities for each type of epilepsy. Further, when properly selected, three network indices suffice to provide a comprehensive description of these differences. Yet, since such uniqueness in the pattern of differences is also evident in the power spectrum, we conclude that the added value of the graph theory approach in this context should not be overestimated.

YNIMG Journal 2014 Journal Article

Guiding functional connectivity estimation by structural connectivity in MEG: an application to discrimination of conditions of mild cognitive impairment

José Angel Pineda-Pardo
Ricardo Bruña
Mark Woolrich
Alberto Marcos
Anna C. Nobre
Fernando Maestú
Diego Vidaurre

Whole brain resting state connectivity is a promising biomarker that might help to obtain an early diagnosis in many neurological diseases, such as dementia. Inferring resting-state connectivity is often based on correlations, which are sensitive to indirect connections, leading to an inaccurate representation of the real backbone of the network. The precision matrix is a better representation for whole brain connectivity, as it considers only direct connections. The network structure can be estimated using the graphical lasso (GL), which achieves sparsity through l1 -regularization on the precision matrix. In this paper, we propose a structural connectivity adaptive version of the GL, where weaker anatomical connections are represented as stronger penalties on the corresponding functional connections. We applied beamformer source reconstruction to the resting state MEG recordings of 81 subjects, where 29 were healthy controls, 22 were single-domain amnestic Mild Cognitive Impaired (MCI), and 30 were multiple-domain amnestic MCI. An atlas-based anatomical parcellation of 66 regions was obtained for each subject, and time series were assigned to each of the regions. The fiber densities between the regions, obtained with deterministic tractography from diffusion-weighted MRI, were used to define the anatomical connectivity. Precision matrices were obtained with the region specific time series in five different frequency bands. We compared our method with the traditional GL and a functional adaptive version of the GL, in terms of log-likelihood and classification accuracies between the three groups. We conclude that introducing an anatomical prior improves the expressivity of the model and, in most cases, leads to a better classification between groups.

YNICL Journal 2014 Journal Article

The Default Mode Network is functionally and structurally disrupted in amnestic mild cognitive impairment — A bimodal MEG–DTI study

Pilar Garcés
José Ángel Pineda-Pardo
Leonides Canuet
Sara Aurtenetxe
Maria Eugenia López
Alberto Marcos
Miguel Yus
Marcos Llanero-Luque

Over the past years, several studies on Mild Cognitive Impairment (MCI) and Alzheimer's disease (AD) have reported Default Mode Network (DMN) deficits. This network is attracting increasing interest in the AD community, as it seems to play an important role in cognitive functioning and in beta amyloid deposition. Attention has been particularly drawn to how different DMN regions are connected using functional or structural connectivity. To this end, most studies have used functional Magnetic Resonance Imaging (fMRI), Positron Emission Tomography (PET) or Diffusion Tensor Imaging (DTI). In this study we evaluated (1) functional connectivity from resting state magnetoencephalography (MEG) and (2) structural connectivity from DTI in 26 MCI patients and 31 age-matched controls. Compared to controls, the DMN in the MCI group was functionally disrupted in the alpha band, while no differences were found for delta, theta, beta and gamma frequency bands. In addition, structural disconnection could be assessed through a decreased fractional anisotropy along tracts connecting different DMN regions. This suggests that the DMN functional and anatomical disconnection could represent a core feature of MCI.

YNIMG Journal 2011 Journal Article

Principles of recovery from traumatic brain injury: Reorganization of functional networks

Nazareth P. Castellanos
Inmaculada Leyva
Javier M. Buldú
Ricardo Bajo
Nuria Paúl
Pablo Cuesta
Victoria E. Ordóñez
Cristina L. Pascua

Recovery after brain injury is an excellent platform to study the mechanism underlying brain plasticity, the reorganization of networks. Do complex network measures capture the physiological and cognitive alterations that occurred after a traumatic brain injury and its recovery? Patients as well as control subjects underwent resting-state MEG recording following injury and after neurorehabilitation. Next, network measures such as network strength, path length, efficiency, clustering and energetic cost were calculated. We show that these parameters restore, in many cases, to control ones after recovery, specifically in delta and alpha bands, and we design a model that gives some hints about how the functional networks modify their weights in the recovery process. Positive correlations between complex network measures and some of the general index of the WAIS-III test were found: changes in delta-based path-length and those in Performance IQ score, and alpha-based normalized global efficiency and Perceptual Organization Index. These results indicate that: 1) the principle of recovery depends on the spectral band, 2) the structure of the functional networks evolves in parallel to brain recovery with correlations with neuropsychological scales, and 3) energetic cost reveals an optimal principle of recovery.

YNIMG Journal 2011 Journal Article

The correlation between white-matter microstructure and the complexity of spontaneous brain activity: A difussion tensor imaging-MEG study

Alberto Fernández
Marcos Ríos-Lago
Daniel Abásolo
Roberto Hornero
Juan Álvarez-Linera
Nuria Paul
Fernando Maestú
Tomás Ortiz

The advent of new signal processing methods, such as non-linear analysis techniques, represents a new perspective which adds further value to brain signals' analysis. Particularly, Lempel–Ziv's Complexity (LZC) has proven to be useful in exploring the complexity of the brain electromagnetic activity. However, an important problem is the lack of knowledge about the physiological determinants of these measures. Although a correlation between complexity and connectivity has been proposed, this hypothesis was never tested in vivo. Thus, the correlation between the microstructure of the anatomic connectivity and the functional complexity of the brain needs to be inspected. In this study we analyzed the correlation between LZC and fractional anisotropy (FA), a scalar quantity derived from diffusion tensors that is particularly useful as an estimate of the functional integrity of myelinated axonal fibers, in a group of sixteen healthy adults (all female, mean age 65. 56±6. 06years, intervals 58–82). Our results showed a positive correlation between FA and LZC scores in regions including clusters in the splenium of the corpus callosum, cingulum, parahipocampal regions and the sagittal stratum. This study supports the notion of a positive correlation between the functional complexity of the brain and the microstructure of its anatomical connectivity. Our investigation proved that a combination of neuroanatomical and neurophysiological techniques may shed some light on the underlying physiological determinants of brain's oscillations.

YNIMG Journal 2010 Journal Article

Oscillatory activity in prefrontal and posterior regions during implicit letter-location binding

Pablo Campo
Claudia Poch
Fabrice B.R. Parmentier
Stephan Moratti
Jane V. Elsley
Nazareth P. Castellanos
José María Ruiz-Vargas
Francisco del Pozo

Many cognitive abilities involve the integration of information from different modalities, a process referred to as “binding. ” It remains less clear, however, whether the creation of bound representations occurs in an involuntary manner, and whether the links between the constituent features of an object are symmetrical. We used magnetoencephalography to investigate whether oscillatory brain activity related to binding processes would be observed in conditions in which participants maintain one feature only (involuntary binding); and whether this activity varies as a function of the feature attended to by participants (binding asymmetry). Participants performed two probe recognition tasks that were identical in terms of their perceptual characteristics and only differed with respect to the instructions given (to memorize either consonants or locations). MEG data were reconstructed using a current source distribution estimation in the classical frequency bands. We observed implicit verbal–spatial binding only when participants successfully maintained the identity of consonants, which was associated with a selective increase in oscillatory activity over prefrontal regions in all frequency bands during the first half of the retention period and accompanied by increased activity in posterior brain regions. The increase in oscillatory activity in prefrontal areas was only observed during the verbal task, which suggests that this activity might be signaling neural processes specifically involved in cross-code binding. Current results are in agreement with proposals suggesting that the prefrontal cortex function as a “pointer” which indexes the features that belong together within an object.

YNIMG Journal 2005 Journal Article

Is medial temporal lobe activation specific for encoding long-term memories?

Pablo Campo
Fernando Maestú
Tomás Ortiz
Almudena Capilla
Santiago Fernández
Alberto Fernández

Several neuroimaging studies have consistently demonstrated the critical involvement of prefrontal cortices and medial temporal lobes during long-term encoding. While the contribution of prefrontal lobes to working memory is well established, the role of the MTL structures remains controversial. To address this issue, we registered the neuromagnetic brain patterns of eight adult volunteers while they performed two working memory tasks (verbal and spatial) using magnetoencephalography (MEG). MEG recordings can provide real-time measures of brain activity, thus allowing detailed tracking of the time-course of brain activation during the encoding phase. We detected sustained and material-specific activity on the MTLs during the encoding phase of a working memory task, based on verbal and spatial information. Two peaks of activation were noted in the left MTL during word encoding in two non-consecutive time periods (500–600 ms and 700–800 ms after stimulus onset). Right MTL laterality was found for encoding locations when we collapsed activity sources in a wider time period (400–800 ms). In addition, we provided the spatiotemporal profiles of what seems to be two different brain circuits specific for each type of material. Finally, following an emerging conceptualization of working memory, we hypothesized that encoding processes mediated by the MTL to long-term memory would also apply to working memory.

YNIMG Journal 2004 Journal Article

Spatiotemporal brain dynamics during preparatory set shifting: MEG evidence

José A Periáñez
Fernando Maestú
Francisco Barceló
Alberto Fernández
Carlos Amo
Tomás Ortiz Alonso

Humans can flexibly alter a plan of action to adjust their behavior adaptively in changing environments. Functional neuroimaging has shown distinct patterns of activation across a frontoparietal network responsible for switching and updating such plans of action or ‘task sets. ’ However, little is known about the temporal order of activations within prefrontal or across with posterior regions subserving set-shifting operations. Here, whole-head magnetoencephalography (MEG) was used to explore the spatiotemporal brain dynamics in a modified version of the Wisconsin card-sorting test (WCST). Our task was designed to examine preparation of set-shifting rather than set-acquisition operations time locked to context-informative cues. Three cortical regions showed a larger number of MEG activity sources in response to shift and relative to nonshift cues: (a) inferior frontal gyrus (IFG; BA 45, 47/12), (b) anterior cingulate cortex (ACC; BA 24, 32), and (c) supramarginal gyrus (SMG; BA 40). Importantly, the timing of MEG activation differed across these regions. The earliest shift-related MEG activations were detected at the IFG (100–300 ms postcue onset), followed by two further peaks at the ACC (200–300 and 400–500 ms) and the SMG (300–400 and 500–600 ms). Several other prefrontal and posterior cortical areas were similarly activated by both shift and nonshift preparatory cues. The resulting temporal pattern of interactions within prefrontal and across with posterior association cortices is coherent with current models of task switching and provides novel information about the temporal course of brain activations responsible for the executive control of attention.

YNIMG Journal 2003 Journal Article

Modulation of brain magnetic activity by different verbal learning strategies

Fernando Maestú
Panagiotis G Simos
Pablo Campo
Alberto Fernández
Carlos Amo
Nuria Paul
Javier González-Marqués
Tomás Ortiz

In this study we examined spatiotemporal profiles of brain activity in the context of tasks designed to engage different verbal learning strategies (serial order, phonological, and semantic). The profile of activation associated with the serial-order strategy, which resulted in poor recall performance, featured early activation of the inferior frontal, sensorimotor, and insular region in the left hemisphere, between 200 and 400 ms after stimulus onset. Subsequently, activation was more prominent in dorsolateral prefrontal cortices bilaterally. In contrast, activation profiles associated with the phonological strategy featured predominantly activation of the superior temporal gyrus in the left hemisphere between 500 and 600 ms. Predominant activation of the left middle temporal gyrus, between 500 and 700 ms, was the key feature of the activation profile observed when the semantic elaboration strategy was utilized. These results suggest that different brain circuits are engaged to support learning of new verbal information as a function of the level and type of initial processing applied to the stimuli.

YNIMG Journal 2002 Journal Article

Spanish Language Mapping Using MEG: A Validation Study

Fernando Maestú
Tomás Ortiz
Alberto Fernandez
Carlos Amo
Pilar Martin
Santiago Fernández
Rafael G. Sola

The purpose of the present study was to compare magnetoencephalography (MEG) data with the results of the intracarotid amytal procedure (IAP). Twenty-one native Spanish-speaking patients with intractable epilepsy underwent MEG language mapping. A subset of 8 patients also underwent an IAP. With the exception of 2 patients who showed right hemisphere dominance, all other patients showed left hemisphere dominance for language on the MEG recording. The IAP findings were consistent with MEG results in 7 patients. The eighth patient who, according to the MEG data, had probable right hemisphere dominance for language did not show clear hemispheric specialization for language on the IAP and suffered a transient global aphasia following a right temporal lobotomy. These results suggests that MEG-based language mapping can play an important role in presurgical clinical evaluation.