Invited Talks

Keynotes:

Tutorials:



Keynote I :
CNN : A Brain-like Computer on a Chip

  1. Speaker : Leon O. Chua

    Leon Chua received his MS (1961) and PhD (1964) degrees from the Massachusetts Institute of Technology and the University of Illinois at Champaign-Urbana, respectively. He became an Assistant Professor of Electrical Engineering at Purdue University in 1964, and was promoted to Associate Professor in 1967. He joined the University of California, Berkeley in 1970, and has been a Professor of Electrical Engineering and Computer Sciences.

    He is the first recipient of the 2005 Gustav Kirchhoff Award, the highest IEEE Technical Field Award for outstanding contributions to the fundamentals of any aspect of electronic circuits and systems that has a long term significance or impact. He was also awarded the prestigious IEEE Neural Networks Pioneer Award in 2000 for his contributions in neural networks. He was elected a Fellow of the IEEE in 1974 and has received many international prizes, including the IEEE Browder J. Thompson Memorial Prize in 1972, the IEEE W. R. G. Baker Prize in 1978, the Frederick Emmons Award in 1974, the M. E. Van Valkenhurg Award in 1995, and again in 1998. He was awarded 7 USA patents and 8 Honorary doctorates (Doctor Honoris Causa) from major European universities and Japan. He is also a recipient of the top 15 cited authors in Engineering award in 2002, chosen from the Current Contents (ISI) database of all cited papers in the engineering disciplines in the citation index from 1991 to October 31, 2001, from all branches of engineering. He was elected a foreign member of the European Academy of Sciences (Academia Europea) in 1997.

  2. Abstract

    A novel class of brain-like information-processing systems called cellular neural networks will be introduced. Like a neural network, it is a large-scale nonlinear analog circuit which processes signals in real time. Like cellular automata, it is made of a massive aggregate of regularly spaced circuit clones, called cells, which communicate with each other directly only through its nearest neighbors. Cellular neural networks share the best features of both worlds; its continuous time feature allows real-time signal processing found wanting in the digital domain and its local interconnection feature makes it tailor made for VLSI implementation. Cellular neural networks are uniquely suited for high-speed parallel signal processing. Many CNN universal chips have been developed to perform sophisticated image processing tasks. By programming a sequence of templates, different kinds of processing are possible. In this lecture, many impressive applications of cellular neural networks will be presented along with its theory.

Keynote II :
Converging architectures for nanoscale cellular processor systems

  1. Speaker : Tamás ROSKA

    Tamás ROSKA received the Diploma in electrical engineering from the Technical University of Budapest in 1964 and the Ph.D. and D.Sc. degrees in Hungary in 1973 and 1982, respectively. He is the Fellow of the IEEE and elected member of four Academies of Sciences in Europe.

    Since 1964 he has held various research positions, since 1982 he has been with the Computer and Automation Research Institute of the Hungarian Academy of Sciences where he is presently head of the Analogic and Neural Computing Research Laboratory. He is also a Professor and Dean of the Faculty of Information Technology at the Pázmány P. Catholic University, Budapest. Professor Roska has taught several courses, presently, he is teaching graduate courses on "Emergent Computations" and "Cellular Neural Networks and Visual Microprocessors". In 1974 and since 1989 in each year, he has been Visiting Scholar at the University of California at Berkeley.

    His research interests: cellular wave computing, info-bionics, cellular neural networks, nonlinear circuit and systems, neural electronic circuits, and analogic spatial-temporal supercomputing and computational complexity. He has published more than hundred research papers and four books (partly as a co-author). His seminal paper on the CNN Universal Machine, co-authored by L. O. Chua, has received several hundred citations.

    Dr. Roska is a co-inventor of the CNN Universal Machine (with Leon O. Chua) and the analogic CNN Bionic Eye (with Frank S. Werblin and Leon O.Chua), both are US patents owned by the University of California at Berkeley.

    In 2002, 2003 he had been serving as Editor-in-Chief of the IEEE Transactions on Circuits and Systems He is a member of the Editorial Board of the International Journal of Circuit Theory and Applications, the Journal of the Franklin Institute, and the Neural Processing Letters. He has been a founding Chair of the Technical Committee on Cellular Neural Networks and Array Computing in the IEEE Circuits and Systems Society. He received the IEEE Third Millenium Medal and the IEEE Circuits and Systems Society's Golden Jubilee Award. He has been awarded a "doctor honoris causa" from the University of Veszprém.

    Dr. Roska received in Hungary the Széchenyi Prize, the Szentgyörgyi Prize and the D. Gabor Prize, the Grand Prize of the "Pro Renovanda Cultura Hingariae", and very recently the 2002 Bolyai Prize.

    Dr. Roska is a member of the Hungarian Academy of Sciences, the Academia Europaea, the European Academy of Arts and Sciences, the St. Steven Academy, and a Fellow of the IEEE.

  2. Abstract

    What will be the prototype computing architecture for the nanoscale era? We can now place one million 8-bit microprocessors on a 45 nm scale CMOS chip. The top supercomputer hosts a quarter million processors. The physical constraints suggest cellular topographic architectures. In this lecture, one answer to this challenge will be reported: the Cellular Wave Computer. The architecture, the various physical implementations with and without integrated sensory arrays, the biological relevance, and the algorithmic principles will be described. Some application highlights will also be presented, including a solution with over 20 000 frame per second input image flow and real time decision at this frame rate.



Tutorial I : Digital image processing and analysis for medical applications

  1. Speaker : Andrzej Materka

    Technical University of Lodz, Institute of Electronics
    Wolczanska 211/215, 90-924 Lodz, Poland
    materka@p.lodz.pl

  2. Abstract

    The aim of his tutorial is to introduce the ISITC ’07 Symposium participants into the field of methods, algorithms and techniques of biomedical image acquisition, processing and analysis. The state of the art of this field will be also presented. To assess the state of human body and its organs, which is necessary for medical diagnosis and/or disease treatment, one needs objective information about physical and chemical properties of various tissues. However, normally there is no direct access to a living tissue. Thus, noninvasive techniques of acquiring quantitative information about parts of the human body are of great value for the task. On the other hand, mainstream of the information that comes to human mind from the external world has the form of visual images and their sequences. It allows most compact and efficient description of objects in complex environment – as a popular saying expresses, “an image is worth more than a thousand words”. These reasons explain why searching for noninvasive imaging techniques is one of the main areas of activity of researchers, with a goal of developing fast, objective and precise means aiding medical doctors in their profession. An excellent example of a success in this field is a revolutionary technique of magnetic resonance imaging – developed in the seventies of the past century and now widely applied to hospitals and clinics. The images contain information that needs to be extracted from the raw data. Some specific models are assumed for the objects represented in an image, and model parameters are computed. Segmentation of the image data has to be performed to identify the objects. Based on the extracted parameters one can classify the objects into different classes (e.g. normal and cancerous tissue), or measure changes of the tissue properties over time (e.g. to monitor state of bones affected by osteoporosis). Prior to the steps of image segmentation and quantitative description (analysis), one often applies preprocessing routines to raw data, e.g. to suppress thermal noise components originated in measurement devices. Another important application is to visualize internal parts of the human body (e.g. blood vessels - venous or arterial trees) for diagnosis, therapy planning and education purposes. Prior to visualization, one needs to build geometrical models of internal organs on the basis of available image information. The validity of various methods and algorithms developed for quantitative analysis of biomedical images have to be verified with the use of test objects – physical and/or digital, and finally accepted by users – experts in the medical field.

    For more detailed information of this tutorial, click here

Tutorial II : Wireless Sensor Networks: From Theory to Practice

  1. Speaker : Martin Haenggi

    Associate Professor, Dept. of Electrical Engineering, University of Notre Dame
    Notre Dame, IN 46556, USA
    mhaenggi@nd.edu

  2. Abstract

    Wireless sensor networks combine distributed sensing, computing, and wireless communications into a powerful technology that offers unprecedented resolution, unobtrusiveness, and autonomous operation for countless applications. At the same time, they offer numerous challenges, in particular the strict energy constraints, the distributed operation, and the scalability. This tutorial provides a comprehensive and self-contained introduction to wireless sensor networks, covering all the relevant aspects from the basic theory to real-world applications. It consists of four parts:

    1. Introduction: Motivation, relevance, and important applications

    2. Challenges and solutions: Difference to other wireless networks; modelling issues; energy-efficient network protocols; performance limits and quality-of-service issues

    3. Practical aspects: Hardware overview; experimental results and measurements.

    4. Conclusions and outlook.

    For more detailed information of this tutorial, click here