Research Unit of Magnetoencephalography
for biomedical applications
The CNR’s Research Unit of Magnetoencephalography for biomedical applications (MEG-BioApp) is located in Naples at the Hermitage Capodimonte Clinic. MEG-BioApp is the result of a scientific agreement among the Institute of Applied Sciences and Intelligent Systems of CNR (ISASI-CNR), the University of Naples Parthenope and Hermitage Capodimonte.
The Unit addresses the interdisciplinary study of neurological syndromes such as preclinical phase of dementia (MCI), Alzheimer’s disease, frontotemporal dementia, Parkinson disease, autism spectrum disorders (ASD), amyotrophic lateral sclerosis and migraine.
The aim is to use magnetoencephalography to 1) select biomarkers to monitor progression of diseases and 2) to foster our understanding of the pathophysiology of neurodegeneration.
Furthermore, ongoing research (based on technologies developed by CNR-ISASI) aims at the optimization of ultra-sensitive magnetic sensors for their application in functional brain imaging.
|Carmine Granata, PhD
Responsible of Research Unit
|Giuseppe Sorrentino, MD, PhD
University of Naples Parthenope
Michele Ambrosanio, Engineer, (University of Naples Parthenope)
Fabio Baselice, Engineer, (University of Naples Parthenope)
Giampaolo Ferraioli, Engineer, (University of Naples Parthenope)
Stefano Franceschini, Engineer, (University of Naples Parthenope)
Anna Lardone, Psychologist, (University of Naples Parthenope)
Marianna Liparoti, Exercise physiologist, (University of Naples Parthenope)
Laura Mandolesi, Psychologist, (University of Naples Federico II)
Roberta Minino, Exercise physiologist, (University of Naples Parthenope)
Matteo Pesoli, Psychologist, (University of Naples Parthenope)
Arianna Polverino, Biologist, (Hermitage Capodimonte Institute)
Rosaria Rucco, Engineer, (University of Naples Parthenope)
Pierpaolo Sorrentino, Neurologist, (University of Naples Parthenope)
Emahnuel Troisi Lopez, Exercise physiologist, (University of Naples Parthenope)
Antonio Vettoliere, Engineer, (ISASI-CNR)
Vincenzo Bonavita, (Univerity of Naples Federico II, Naples, Italy)
Michael Breakspear, (University of Newcastle, NSW, Australia)
Luca Cocchi, (QIMR Berghofer, Brisbane, QLD, Australia)
Matteo Demuru, (SEIN-Stichting Epilepsie Instellingen, Amsterdam, The Netherlands)
Anna Maria D’Ursi (University of Salerno, Salerno, Italy)
Matteo Fraschini, (University of Cagliari, Cagliari Italy)
Antonio Fratini, ( School of Life and Health Sciences, Aston University, Birmingham, UK)
Nathalie George, (Social and Affective Neuroscience Lab, Paris, France)
Leonardo Gollo, (QIMR Berghofer, Brisbane, QLD, Australia)
Arjan Hillebrand, (VU University Medical Centre, Amsterdam, The Netherlands)
Viktor Jirsa, (Institut de Neurosciences des Systèmes, Marseille, France)
Vittorio Pizzella, (University of Chieti-Pescara, Chieti, Italy)
Mario Quarantelli, (Institute of Biostructures and Bioimaging of CNR, Naples, Italy)
Maria Luisa Scattone, (Istituto Superiore di Sanità, Rome, Italy)
Denis Schwartz, (CERMEP, Lyon, France)
Cornelis J. Stam, (VU University Medical Centre, Amsterdam, The Netherlands)
Franca Tecchio, (Institute of Cognitive Sciences of CNR, Rome, Italy)
Andrew Zalesky, (University of Melbourne, Melbourne, VIC, Australia)
Who we are
We are a group of young researchers with various backgrounds, including physicists, neurologists, computer scientists, electronic engineers, psychologists, movement scientists. We are truly committed to multidisciplinary work, and make great effort to overcome background differences to tackle problems from new perspectives. We like brain–storming, trying out new things, and high–risk innovative research. Although our work on the clinical applications of MEG began just two years ago, we grew rapidly and are eager to engage in research in many more aspects of the brain’s physiological and pathological activity.
What our commitment is
According to our story, we strongly believe that “contamination” is a source of enrichment in science. Are you a scientist interested in the new frontiers of neurosciences? Do you have an idea and want to try it out on real data? Do you have a clinical hypothesis but lack the technical skills to set up experiments or analyse data? Do you simply want to show your ideas?
We are looking for collaborations and can’t wait to share ideas and projects. Contact us, we will be very happy to meet you!!
Where we come from
The MEG system was developed by ISASI-CNR in the framework of a research project funded by MIUR. Following a research agreement between the ISASI-CNR, the Parthenope University (Naples) and the Hermitage Capodimonte Clinic (Naples), the MEG started to work in the Hermitage clinical environment since the beginning of 2014.
What is MEG
Magnetoencephalography (MEG) is a non-invasive neurophysiologic technique that measures the magnetic fields generated by the neuronal activity using ultra high sensitivity magnetic sensors (superconducting quantum inference devices, SQUIDs). MEG is a purely passive method, i.e. a completely non-invasive measurement requiring no contrast agent, magnetic field or x-ray. MEG allows direct measurement of neural oscillations. The extremely high temporal resolution of MEG allows for an assessment of brain activity on a timescale not accessible to other brain functional imaging methods, i.e. Single-Photon Emission Computed Tomography (SPECT), Positron Emission Tomography (PET), or functional Magnetic Resonance Imaging (fMRI). This is particularly important given the evidence that neural oscillations might represent an intrinsic physiological process by which communication among neurons occurs. Compared to Electroencephalography (EEG), MEG offers a better source localization, due to the reduced signal distortion. In fact, MEG measurements are not distorted or attenuated by the insulating layers such as the skull, tissues or anatomical open spaces as in the EEG.
Our MEG system
The system, developed by ISASI-CNR, is based on an array of 154 ultra high sensitive SQUID magnetometers arranged in a helmet shape to cover entirely the subject’s head. Further, nine SQUID sensors are arranged in a three vector magnetometers acting as references and located out of measurement plane. This geometry requires a careful wiring and the SQUID sensors have been designed to minimize crosstalk.
The sensorial helmet are immerged in helium bath thanks to high performance fiberglass Dewar having a capacity of 74 l and a magnetic field noise at most comparable the SQUID sensors one. The low heat transfer guarantees up to seven days of measurement with one refill of helium. Since the magnetic signals generated by the human brain are at least eight orders of magnitude smaller than the magnetic disturbances, the MEG system is housed in a magnetically-shielded room (MSR), which drastically reduces environmental noise. The MSR is made of aluminium and µ-metal nested layers to reduce, respectively, high- and low-frequency noises. The inner dimensions define a room of 2.9 × 3.7 × 2.9 m. All the electric connections have been designed to block any magnetic noise propagating through the connection wires. The shielding factor exhibited by MRS is 60 dB at 1 Hz and 100 dB at 50 Hz.
The MEG system is also equipped with a 32 integrated non-magnetic EEG-channels cap, with ultra-thin wires, which is optimal for usage inside a MEG helmet. This system allows data to be stored digitally and analysed using both commercial and open-source programs. Scalp EEGs can be inspected visually in real time. It is also possible to record electrooculograms (EOGs) and electrocardiograms (ECGs) using additional electrodes. Moreover, visual and auditory stimuli and seven external triggers can be used.
Source reconstruction research
The magnetic fields are recorded outside of the head and then analyzed to localize the sources of the activity within the brain. This is an ill–posed problem, and there are a number of techniques that differ in various regards (for instance in the assumptions used to introduce constraints). In our group, we use a technique called beamformer. Furthermore, we have a team dedicated to the development of newer beamformer algorithms, in order to tailor the source reconstruction to the specific experimental paradigm. The location of the sources can be superimposed on anatomical images, such as MRI (i.e. either the native MRI or a template MRI) to provide information about both structure and function of the brain. MEG possess both a good spatial and an excellent temporal resolution, thus allowing the investigation of both physiological and pathological processes in neurology.
The MEG recording of brain magnetic activity enables the investigation of neuronal activity such as cognitive processes, language perception, memory encoding and retrieval and higher-level tasks.
MEG is a useful tool for the identification of early diagnostic biomarker in numerous diseases, as it allows the study of brain functions and facilitates the investigation of both spontaneous and evoked activities. Nowadays, MEG is used in some clinical conditions such as preoperative assessment of brain tumours and intractable epilepsy (pre-surgical mapping). Furthermore, MEG is useful to investigate neurological conditions such as neurodegenerative diseases, multiple sclerosis, amyotrophic lateral sclerosis and migraine, as well as to investigate the autism spectrum disorders and the functional recovery after stroke.
The connectivity – “an odd kind of sympathy”
The idea that physical laws govern the synchronization of oscillators dates back to the pioneering work of Christiaan Huygens (1629–1695), that realized in his work “Horologium Oscillatorium sive de motu pendulorum” and in a paper entitled “An odd kind of sympathy” that two coupled oscillators could show the tendency to synchronize. Many steps forward have been done since, and in a long process, this concept landed in neuroscience, when it became clear that some properties of the neurons could be predicted by modelling them as oscillators. We apply different connectivity metrics and study their properties, in order to detect what is their behaviour and which ones can be applied successfully and meaningfully to predict clinical outcomes. In order to do this, we use freely available software as well as the code implemented by our team.
Network theory – from bench to bedside
Our brain can be modelled as an ensemble of parts that have all sorts of non linear interactions among themselves. Hence, the graph theory lends itself nicely to capture properties of the brain that are derived by simple rules applied to each of its elements (i.e. emerging properties). Indeed, network theory provides us with a powerful mathematical framework to study all the complex non linear interactions among multiple brain areas in a way that allows interpretable results. Two major discoveries, the existence of the small world networks and the scale free networks, paved the way for the application of network metrics to the study of the brain. We are now engaged in the applications of known network metrics to measure topological differences between a range of neurodegenerative diseases, as well as to monitor those properties longitudinally during the natural history of diseases. Furthermore, we engage in the application of measures capturing the resiliency of networks in neurodegeneration and in the healthy brain, as well as to find ways to adapt existing measures to the specific properties of MEG signals.
Ongoing research projects
- Brain networks topology in:
- Cognitive disorders
- Parkinson’s disease
- Amyotrophic Lateral Sclerosis
- Depressive disorder
- Autism Spectrum Disorders
MEG and cognitive disorders
The pathogenesis of neurodegenerative disorders, such as Alzheimer’s disease (AD) or frontotemporal degeneration (FTD), is still poorly understood and the exact correlation between pathological findings, (such as amyloid plaques and neurofibrillary tangles), and the neurodegenerative process is unclear. Moreover the large heterogeneity of the clinical presentations of these diseases has no unambiguous explanation. In connectomics, AD is described as a “disconnection syndrome”, and the evidence of a correspondence between hubs and regions most susceptible to amyloid deposition suggests a greater vulnerability to AD insults by the most active areas, with a higher basal metabolism. The aim of our current work is to exploit these properties in order to improve early classification of patients.
MEG and PD
Parkinson’s disease is the most common basal ganglia disorders. As of today, the aetiology of Parkinson’s disease is still unknown and the diagnosis remains a clinical one. With our MEG(based on magnetometers) we aim at the reconstruction of the activity of the basal ganglia. This is especially interesting, since correlations between α-synuclein and altered neuronal activity have been described. In this framework, we explore how the different cross-talk among brain areas underlies the symptoms and/or the therapeutic response.
MEG and ALS
Amyotrophic lateral sclerosis is one of the most devastating brain disorders. While classically considered to be targeting the motor neurons, recently it has become evident that a much more complex pathophysiological mechanism underlies ALS. Most interestingly, an association between ALS and frontotemporal dementia has been described. We would like to know better how the whole brain evolves as the disease progresses. Hence, we perform network analysis on MEG recording of ALS patients at various stages of the disease.
MEG and migraine
Migraine affects roughly 15% of the world population. We aim at understanding how the same stimulus is processed differently in the brain of migraine patients as compared to controls. Indeed, cortical hyperexcitability has been described as one of the hallmarks of migraine. We used a median nerve stimulation paradigm in order to try to quantify this phenomenon.
MEG and depression
According to the World Health Organization, depression affects about 350 million people worldwide. The diagnosis of depression is based on clinical criteria, since we lack a clear definition of the pathophysiological mechanisms leading to this condition. Furthermore, the clinical responsiveness to treatment is assessed empirically. Recently, it has been showed that the activity of the prefrontal cortex is altered in depressive disorders. Furthermore, complexity analysis applied to MEG was shown to identify patients from controls. By recording MEG before and after treatment, we hope to identify connectivity markers of responsiveness to treatment.
MEG and Autism Spectrum Disorders
Autism Spectrum Disorders (ASD) are defined by restricted/stereotyped behaviours and social impairments and have a massive impact on both the individuals with ASD and society itself. The biological basis are unknown and a fully effective treatment has yet to be realized. Given the neurodevelopmental basis of ASD, MEG emerges as an important investigatory tool to explore novel early biomarkers useful not only for diagnostic and prognostic purposes, but also for stratification and response indices for treatment development.
Institute of Applied Sciences and Intelligent System
Unita Di Ricerca Territoriale CNR
University Of Naples Parthenope
c/o Hermitage-Capodimonte Clinic
Via cupa delle tozzole 2 – 80131 Naples
Get in touch:
www.isasi.cnr.it/imeg – napolimeg @ gmail.com