Tutorials

The 15th IEEE International Symposium on Signal Processing and Information Technology

ISSPIT 2015



December 7-10, 2015, Abu Dhabi, UAE

Tutorials

Stochastic Modeling for Medical Image Analysis




Tutorial Organizers

Ayman El-Baz

BioImaging Laboratory, Bioengineering Department, University of Louisville, Louisville, KY 40292, USA.

Email: aselba01@exchange.louisville.edu


Georgy Gimel'farb

Department of Computer Science, University of Auckland, Auckland 1142, New Zealand.

Email: g.gimelfarb@auckland.ac.nz

Tutorial Presenters-Contributors

Ayman El-Baz, Ahmed Elnakib, Fahmi Khalifa, Matthew Nitzken, and Ahmed Soliman

BioImaging Laboratory, Bioengineering Department, University of Louisville, Louisville, KY 40292, USA,

Email: {aselba01,aaelna02,fakhal01,mjnitz02,asnaee01} @exchange.louisville. edu


Georgy Gimel'farb

Department of Computer Science, University of Auckland, Auckland 1142, New Zealand.

Email: g.gimelfarb@auckland.ac.nz


Outline

Stochastic modeling facilitates understanding of natural phenomena depicted in medical anatomical and functional (dynamic) MRI and CT images. Computer-assisted diagnostics calls for fast and accurate unsupervised learning of models from images. The tutorial details efficient stochastic modeling techniques, including (i) shape models of objects-of-interest; (ii) shape and visual appearance models based on analytic learning of 2nd- or higher-order non-parametric Markov-Gibbs random fields, and (iii) appearance models based on precise unsupervised learning of a mixture of pseudo-distributions approximating an empirical marginal probability distribution of pixel/voxel intensities. The pseudo-distribution, one per object associated with a prominent mode of the empirical distribution, is a linear combination of unimodal distributions, e.g. discrete Gaussians, with a dominant positive and several sign-alternate subordinate components. Integration of the models and learning techniques will be illustrated in application to early detection of lung and prostate cancer, kidney transplant rejection, and autism, as well as to cardiac functionality assessment.


Covered Topics

  • - Theory of Stochastic Modeling for Medical Image Analysis
  • - Stochastic deformable model (parametric deformable model/level sets)
  • - Vascular system segmentation
  • - Brain segmentation
  • - Lung segmentation
  • - Computer Assisted Diagnosis of acute renal rejection.
  • - Motion analysis of left ventricle wall

Tutorial Format Plan

Half-day tutorial (3 academic hours): 4 lectures on image modeling, learning techniques, and solutions of diagnostic problems with demonstrations of software packages and experimental results. All the attendees will receive handouts in the electronic form (pdf and/or ppt files).

Anticipated Number of Participants

The anticipated number of participants is within 20 – 40.

Design and Implementation of Advanced Video imaging systems



Tutorial Presenters

Prof. Abbes Amira

Prof. Abbes Amira received his Ph.D. in the area of Computer Engineering from Queen’s University Belfast in 2001, United Kingdom, developing a coprocessor for matrix algorithms using reconfigurable computing for image and signal processing applications. Prof. Amira took academic and consultancy positions in the United Kingdom and overseas, including his current positions as a Professor in Computer Engineering at Qatar University, Qatar and a full professor in visual communications and leader of the Visual Communication Cluster at the University of the West of Scotland (UWS), UK. He took other academic positions at the University of Ulster-UK, Qatar University-Qatar, Brunel University-UK and Queen’s University Belfast-UK. He has been awarded a number of grants from government and industry and has published more than 200 publications in the area of reconfigurable computing, image and signal processing during his career to date. He has successfully supervised 13 PhD students under his supervision in the area of image processing and reconfigurable computing.He has been invited to give talks, short courses and tutorials at universities and international conferences and being chair, program committee for a number of conferences. He was one of the tutorials presenters at IEEE International Conference on Image Processing ICIP 2009, Chair of ECVW 2011, Program Chair of ECVW2010, Program Co-Chair of ICM12, DELTA 2008, IMVIP 2005. He is the general conference Co-Chair for ICM 2014. He contributed as program committee member to many conferences in the areas of image and signal processing, reconfigurable computing, embedded systems and connected health. He a member of the IEEE Technical Committee for Biomedical Circuits and Systems. He is also one of the 2008 VARIAN prize recipients. He has been a PhD external examiner for Trinity College Dublin, Edinburgh, Southampton and many other Universities in UK, Hong Kong, Australia and Finland. Prof. Amira was one of the guest editors for the Special Issue in the Pattern Recognition Journal, titled ”Feature Generation and Machine Learning for Robust Multimodal Biometrics”, March 2008. He took consultancy positions with many companies in UK, and he holds a visiting professor position at University of Tunn Hussein Onn, Malaysia. Previously, he was visiting Professor with the University of Nancy, Henri Poincare, France. He is a regular referee for many national and international funding bodies, including (EPSRC and QNRF) and he sits on the international advisory boards for some international research centers. He is a Fellow of IET, Fellow of the Higher Education Academy, Senior member of the IEEE, and Senior member of ACM. His research interests include: embedded systems, high performance reconfigurable computing, image and video processing, multi-resolution analysis, biometrics and connected health applications.


Dr. Naeem Ramzan

Dr. Naeem Ramzan (S’04-M’08-SM’13) has over 10 years of experiences in the field of video processing and communications. Before joining University of West of Scotland (UWS), he was senior fellow researcher and Lecturer (Assistant Prof.) at Queen Mary University of London. He is currently a Reader in Visual Communication in UWS. He has investigated/co-investigated/participated several projects funded by EU and UK research councils. Dr Ramzan received his M.Sc. in telecommunication from University of Brest, France and PhD in electronic engineering from Queen Mary University of London in 2004 and 2008 respectively. He has authored or co-authored of over 80 research publications including journals, book chapters and standardised contributions. He edited the book on “Social Media Retrieval” in Springer, 2013. He is Co-Editor-in-Chief of the VQEG E-letter. I served as Guest Editor in Signal Processing Image Communication Journal and IEEE COMSOC MMTC E-letter. He organised and co-chaired of three ACM Multimedia workshops and served as session chair/co-chair in number of conferences including IEEE ICM, IEEE ICECS and ACM Multimedia. He has been invited to give talks and tutorial in different UK universities. He has served the programme committee member of number of conferences including ICIP from last five years. He has been part of standardisation bodies like MPEG, and VQEG. He is a Co- chair of MPEG High Efficiency Video Coding (HEVC/H265) Verification Group (AHG5), Co-chair of VQEG Ultra High Definition group and a member of British Standard Institution (BSI). He is Senior Member of IEEE, fellow of Higher Education Academy, UK, member of IET, member of IEEE Comsoc and IEEE signal processing society. His research interest includes Image/Video Procession, 2D/3D Video Transmission, Social Media Distribution, Quality of Experience of 2D/3D videos with sensory devices, Heath monitoring environment, and Surveillance centric processing.


Outline

This tutorial will focus on the design and implementation of imaging systems on field programmable gate arrays (FPGAs). We will address different software and hardware issues related to the implementation of algorithms used in applications such as biometrics, medical imaging and video coding. Hardware compilation, partial dynamic reconfiguration and design partitioning for reconfigurable hardware will be also covered, looking at different aspects of: optimisation, code generation and configuration. The tutorial will review the latest FPGA technologies and development methods for imaging systems, including the Zynq System on Chip (SoC) platform, and will conclude with comprehensive case studies demonstrating the deployment of low power reconfigurable architectures for algorithms acceleration and performance evaluation methods for embedded imaging systems. Emphasis will be also on the latest video coding techniques. Different main blocks of H.265/MPEG-HEVC, referred to as High Efficiency Video Coding (HEVC) along with their complexities will be detailed. The potential of parallelism and complexities reduction methodologies will be explained. In addition, the video quality enhancement techniques will be explained in the context of Quality of experience.


Covered Topics

  • - Introduction to Imaging Systems
  • - Video Coding
  • - Reconfigurable Computing for Imaging Systems
  • - FPGA Design Methodologies and Optimization Techniques
  • - Examples and Demos
  • - Conclusions, Q&A

Tutorial Format Plan

Power point (PPT) slides, videos and demos will be presented in this tutorial to illustrate different steps for the design and implementation of imaging systems on reconfigurable hardware.

Anticipated Number of Participants

The anticipated number of participants is within 30 – 40.


Find out more here

Privacy-preserving signal processing



Tutorial Presenters

Michail Maniatakos

Michail Maniatakos is an Assistant Professor of Electrical and Computer Engineering at New York University Abu Dhabi and a Research Assistant Professor at the NYU Polytechnic School of Engineering. Prof. Maniatakos received a B.Sc. and M.Sc. in Computer Science and Embedded Systems from the University of Piraeus, Greece in 2006 and 2007 respectively, as well as an M.Sc., M.Phil. and Ph.D. from Yale University in 2008, 2009 and 2012 respectively. He is the director of the Modern Microprocessor Architectures (MoMA) lab in NYU Abu Dhabi, an IEEE Member and an author of multiple publications in IEEE Transactions and conference papers. He has received the IEEE TTTC Gerald W. Gordon Award for exceptional service to the community, and has won the Hardware Security Embedded System Challenge held at CSAW VIII and CSAW X, in 2011 and 2013 respectively. His research interests include homomorphically encrypted general-purpose computation, hardware security and industrial control systems security.



Outline

Outsourcing computation to the cloud has recently become a very attractive option for enterprises and consumers, due mostly to reduced cost and extensive scalability. At the same time, however, concerns about the privacy of the data entrusted to cloud providers keeps rising. Specific signal processing applications, such as biometric techniques or social network data aggregation, have explicit privacy requirements. Thus, in order to protect the privacy of the data entrusted to third-parties, user opt to employ privacy-preserving algorithms. In this tutorial, we will define privacy and present several applications where privacy is of paramount importance. Then, the discussion will focus on existing methodologies to enable privacy-preserving computation, along with challenges, overhead and implementation discussion. Several cryptography, security and privacy concepts will be presented in the context of signal processing. The tutorial will conclude with real world test cases of privacy-preserving signal processing.


IMPORTANT DATES (2015)
September 4 Proposals for Tutorials and Special Sessions
October 1 Regular paper submission
October 30 Notification of acceptance
November 15 Final Version of paper and registration