Aurel Vasile Martiniuc

Early Stage Researcher

Technische Universität München, Munich , Germany

---------------------------------------------------------------------------------------------------------------------------------------

Dipartimento di Ingegneria Biofisica ed Electtronica, Genoa, Italy

Pieter Laurens Baljon has a background in Artificial Intelligence and pattern recognition. For his Master's thesis at University of Groningen , the Netherlands , he used a Hidden-Markov Model to classify single-trial EEG data in a Brain-Computer Interface.

In the Neuroversity project Pieter Laurens Baljon studies the encoding and decoding of information in networks of dissociated cortical neurons. These networks make it possible to investigate neuronal activity in the absence of specialized structure such as micro circuits. Pieter Laurens Baljon has performed experiments investigating the ability of these networks to discern different input patterns provided via electrical stimulation. Currently he is working on the way the networks integrate multiple inputs. This is an important issue both for engineering embodied neuronal networks, and for understanding real-life phenomena such as attention and sensory integration.

The final part of his PhD research will place these results in the context of more biologically-realistic neuronal preparations. These preparations include neuronal networks with sub populations and ex-vivo networks.

---------------------------------------------------------------------------------------------------------------------------------------

Vladimir Novakov ic

---------------------------------------------------------------------------------------------------------------------------------------

Università degli Studi di Genova, Dipartimento di Informatica, Sistemistica e Telematica, Genoa, Italy

---------------------------------------------------------------------------------------------------------------------------------------

Technische Universität München, Munich , Germany

B. Bhattacharya has been working on implementation of a 2-D neural field model with a view to study closed loop stimulation of the primary motor cortex as a novel therapy for alleviation of pathological tremor observed in Parkinson Disease (PD) patients. A 1-D neural field equation is adapted to represent a 2-D grid of neural mass. This is to emulate the penetration of the primary motor cortex by a 2-D electrode array.

The coding for the work was done in Matlab.

The first step in her work was to implement the 2-D neural field model, coding being done in Matlab. I also built a GUI simultaneously to view the results. The 2-D model was then validated and parameterized against the existing 1-D model. The next step was to study methods and ways to extract the 10Hz frequency component from the voltage signal with progressing time and reconstruct the signal corresponding to this frequency. This reconstructed signal is being used as one of the input parameters to a ‘closed-loop-control term' to attenuate the unwanted frequency of 10Hz observed in the primary motor cortex corresponding to pathological tremor in PD patients.

B. Batthacharva is now a researcher in Intelligent Systems Research Centre at the University of Ulster at Magee (Ireland).

---------------------------------------------------------------------------------------------------------------------------------------

Ecole Nationale Superieure d'Electronique, Informatique et Radiocommunications de Bordeaux, Bordeaux , France

Bilel Belhadj Mohamed was born in Tunis , Tunisia in 1981. He received his masters degree in telecommunications from the High School of Telecommunication in Tunis (SUP'COM) in 2006. During his masters' thesis, he contributed to the development of a Multiprocessor System on Chip (MPSoC) design flow. Currently, he is a 3 rd year Ph.D. student in the “Engineering of Neuromorphic System” group in the ims laboratory (Laboratoire de l'Intégration du Matériau au Système) at the University of Bordeaux 1, France .

His research activities revolve around the development of a biologically plausible hardware simulator for spiking neural networks. The simulator will make it possible to map biophysical models of perception and information processing in biological systems onto VLSI circuits.

Together with other members of the team, he is developing a multi-board multi-chip system with the ability to process neural activity on a 400 neurons-wide network at biologically realistic speeds. Models neurons use the Hodgkin-Huxley model to integrate voltage variability of ionic currents. Synaptic strengths are calculated using a realistic STDP model. Other features like axonal delay and synaptic noise, are also supported.

---------------------------------------------------------------------------------------------------------------------------------------

Ecole Nationale Superieure d'Electronique, Informatique et Radiocommunications de Bordeaux, Bordeaux , France

Yannick Bornat was born in france in 1977. He received his Engineering degree in Electronics and Integrated Circuits and Systems from ENSEIRB and his Masters degree in Microelectronics from the University of Bordeaux, France. He completed his masters degree in 2003. During his degree studies, his main interests were in analog electronics, integrated electronics, low-level digital electronics, processors and digital systems.

For his PhD thesis, he constructed a hardware simulation platform providing very accurate models of adaptive neural networks. The main focus of the work was on analog solutions for differential equations, realtime processing of complex functions in FPGAs, realtime communication and realtime computing.

After a period at IMT (Institute of MicroTechnology - University of Neuchâtel - Switzerland ) where he focused on realtime tools applied to large-scale acquisition systems for living neural networks, he became an ESR at ENSEIRB he worked on signal conditionning, artifact removal and preprocessing for neural measurements.

Today Yannick Bornat is an assistant-professor at IPB in Bordeaux ( France ).

---------------------------------------------------------------------------------------------------------------------------------------

Ecole Nationale Superieure d'Electronique, Informatique et Radiocommunications de Bordeaux, Bordeaux , France

In Neurovers-IT Dr Fabrice Morin was an Experienced Researcher at ENSEIRB where he worked on the development of dedicated hardware for the simulation of neural networks

He is currently a senior researcher and project manager at the Fatronik Foundation (San Sebastian, Spain). He possesses an extensive background in Bioengineering, Neuroscience and Materials Science. Fabrice MORIN has served as a reviewer for the Journal of Micro-ElectroMechanical Systems (IEEE) and Journal of the Royal Society Interface (Royal Society Publishing). He serves as a review editor for the journal Frontiers in Neuroengineering. He has contributed 11 peer reviewed articles (in journals and conference proceedings) on the design and use of microelectrode arrays and other micro-electromechanical systems for biological applications.

---------------------------------------------------------------------------------------------------------------------------------------

University of Twente, Institute for Biomedical Technology, Department Electrical Engineering, Enschede, the Netherlands

Taras Gritsun won his first MSc in 2003, in Vladivostok , Russia , where he worked on remote control systems for a laboratory at the department of electrical engineering. This was followed in 2006 by a second MSc in Lubeck , Germany , where he worked on an experimental setup for simultaneous deep brain stimulation and recording in awake rats. He is currently studying for his Ph.D. at Institute for Biomedical Technology at the University of Twente , where he is part of the the Biomedical Signal and System group, with a fellowship from the Neurovers-IT project.

The focus of his research is on computer modeling of biological neural networks and more specifically on models of the way information is encoded, processed, and learned. The long term goal of the work is to open the way to the development of neuronal controllers for robots and other artificial systems.

---------------------------------------------------------------------------------------------------------------------------------------

Experienced Researcher

University of Twente, Institute for Biomedical Technology, Department Electrical Engineering, Enschede, the Netherlands

Dr Irina Stoyanova has a degree in General Medicine, from the Medical Academy , Sofia , Bulgaria and a PH.D. in neuroscience, from Trakia University , Stara Zagora , Bulgaria . Her research has focused on sensory and autonomic innervation of the gastrointestinal tract, the trigeminal sensory system, and brain control of feeding behavior. She has published over 120 journal and conference papers on the above mentioned topics. She is a member of the Bulgarian and German Anatomical Society, the Bulgarian Society of Neuroscience, and the New York Academy of Sciences. She is also member of the editorial board of International Journal of Biomedical Sciences .She is currently working in Neurovers-IT as a member of the BSS team coordinated by Prof. Wim Rutten, Her current research focuses on chemical coding of cortical neurons and control of their activity.

---------------------------------------------------------------------------------------------------------------------------------------

Vrije Universiteit, Amsterdam, the Netherlands

Oscar Avella's research focuses on neural modeling.

Experimental observations in vivo and in vitro have reported modulation of cortical oscillations, as phases of high synchronization (Waxing) followed by periods of reduced synchronization (Waning). Although the phenomenon is present in almost all frequency bands, it remains not completely understood how this is driven.

Oscar Avella has used the NEURON simulator, to model a network of N e excitatory (E) and N i inhibitory (I) mono compartment cells including leakage, action potential generating Na + and K + channels and network connectivity through AMPA (excitatory) and GABA (inhibitory) synapses.

Under the supervision of Arjen van Ooyen and Jaap van Pelt at the Integrative Neurophysiology department he is using the model to determine whether this network can generate Waxing and Waning and to investigate the role of external background activity.

Partial results show that waxing and waning can be produced by combining specific background spiking input and current baselines delivered to the cells of each population independently. They demonstrate furthermore that the properties of these inputs strongly determine the duration, intensity and frequency of occurrence of waning periods in stable oscillatory networks.

---------------------------------------------------------------------------------------------------------------------------------------

University of Neuchatel , Institute of Microtechnology , Neuchatel , Switzerland

Guilherme Bontorin was born in Curitiba , Brazil , in 1984. He received his Engineering degree in Electronics and Integrated Circuits and Systems from ENSEIRB and the Master degree in Microelectronics from the University of Bordeaux, France in 2006. During this period, his main research interests were in power electronics, power distribution systems, wireless Networks, and PLC communications;

He is currently preparing his PhD degree in Microtechnology at the Institute of Microengineering at the University of Neuchatel , which has recently come under the control of Ecole Polytechnique Féderale de Lausanne (EPFL). The degree will be awarded jointly by University of Neuchatel and ENSEIRB.

he Neurovers-IT project is funding a part of his research. The aim is to develop an enhanced MEA based platform with real time capability to access complex neuronal networks in a closed loop. The final system will be used by others partners in the Neurovers-IT consortium. It includes: a hybrid (living-artificial) neural network, with bidirectional communication (acquisition of signals from and electrical stimulation of the living neural culture), real-time data processing (spike and burst detection, stimulation pattern shaping and triggering,…), and high density and number of acquisition signals (more than 4000).

---------------------------------------------------------------------------------------------------------------------------------------