Special Sessions

Accepted Special Sessions:


    Scope and motivation:
    Computational intelligence comprises a variety of bio-inspired computing paradigms to address complex problems of the real world. Examples of these computing paradigms include, but not limited to, artificial neural networks, fuzzy systems, and evolutionary algorithms. Nevertheless, each intelligent paradigm has its own advantages and disadvantages. Recent advances in the area of computational intelligence have focused on integrating two or more intelligent paradigms together in a common framework in order to harvest their advantages and, at the same time, to minimize their disadvantages, hence the so called “Hybrid Intelligent Systems”.

    Hybrid intelligent systems have been applied successfully to solving a variety of biomedical and bioengineering problems. Examples include the use of intelligent decision support system for medical prognosis and diagnosis. Indeed, an intelligent and reliable decision support system is able to assist medical practitioners in making fast and accurate prognoses and diagnoses, hence saving costs and lives.
    This special session provides a platform for researchers to present and discuss recent advances in hybrid intelligent systems and their application to the biomedical area. The intelligent systems and techniques that are of interest to this special session include, but are not limited to, artificial neural networks, fuzzy set theory and fuzzy systems, genetic and evolutionary algorithms, swarm intelligence, rough sets, chaos theory and chaotic systems. The application areas include, but are again not limited to, simulation of biomedical systems, clinical record and data processing, analysis and synthesis of medical data, medical diagnosis and prognosis, medical image analysis, advanced prosthetics, biosensors, telemedicine, medical informatics and bioinformatics.


    Scope and motivation:
    The new era of data and information fusion in biomedicine is rapidly approaching. Institutions can no longer use paper medical records and are required by law to use electronic medical records resulting in more data. In addition, as new nano‐bio chips (NBCs) make their way into clinical use, there will be an ever‐increasing deluge of molecular and genomic information available for new markers of early detection, markers of pathway heterogeneity, and markers of therapy. Making sense of this rich mixture of data from disparate sources remains one of the greatest challenges for advancing medicine. Expert and clinical decision systems needed for large‐scale data integration should be developed using flexible methods based on novel machine learning and computational intelligence techniques. Novel nonlinear neural adaptive methods can solve difficult non‐linear problems with complex decision boundaries, better fit multimodal functions with landscapes having many local maxima, exploit noise, and yield greater performance when inferior methods breakdown. The enterprise systems sought should employ techniques for unsupervised class discovery (cluster analysis), supervised class prediction, function approximation and signal processing, image and multimedia analysis, text mining, sample and feature dimensional reduction, and data integration with modular and ensemble systems having maximum diversity with minimum shared error. Once developed, these enterprise systems will need to be implemented and results interpreted by teams of scientists and practitioners from many domain areas. Examples include molecular and systems biologists, geneticists, computer scientists and medical informaticians, bioinformaticians, biometricians, biostatisticians, and clinicians specializing in biochemistry, pathology, immunology, endocrinology, neurology, oncology, cardiology, radiology, etc. This session will explore new methods developed for the above topics with a particular focus on machine learning and computational intelligence. Areas of focus include novel machine learning techniques for distance metrics, self‐organizing maps, cluster analysis, classification analysis, function approximation, text mining, image analysis, signal processing, predictive analytics for risk and disease factors, and survival analysis. This session will bring together leading researchers in academics and industry to harness expertise and rapidly thrust forward development of new of expert systems and decision processing. The contribution and public health impact of this special session is to introduce new machine learning and computational methods for biomedical enterprise systems.

    We encourage the submissions of papers on novel machine learning and computational intelligence technologies focusing on biomedicine, with the following technical components:

    • - Feature and sample pre‐processing
    • - Signal processing
    • - Dimension reduction
    • - Text and multimedia mining
    • - Unsupervised neural adaptive learning
    • - Distance metrics
    • - Signal processing
    • - Classification analysis
    • - Function approximation
    • - Electrophysiological signal processing
    • - Integration of molecular data (genomic, proteomic) and clinical data
    • - Classifier fusion and ensemble methods
    • - Knowledge management and organization
    • - Ontology design
    • - Decision support and expert systems

    • - Computer scientists
    • - Bioinformaticians
    • - Nueroinformaticions
    • - Medical informaticians
    • - Statisticians
    • - Molecular biologists
    • - Biomedical and electrical engineers
    • - Other researchers and developers

    • - Leif Peterson,
      The Methodist Hospital Research Institute, Houston, TX USA
      (Email: leifepeterson@sbcglobal.net)

    • - Riccardo Rizzo,
      National Research Council of Italy (CNR), Institute for high performance computing and networking, Palermo, Italy
      (Email: ricrizzo@pa.icar.cnr.it)

    • - Alfonso Maurizio Urso,
      National Research Council of Italy (CNR), Institute for high performance computing and networking, Palermo, Italy
      (Email: urso@pa.icar.cnr.it)


    This special session is organized by the Bioinformatics and Bioengineering Technical Committee (BBTC) of the IEEE Computational Intelligence Society (CIS).

    Scope and motivation:
    Microarray data analysis is an important topic in bioinformatics and computational biology. As genes can be monitored synchronically by microarray technique, microarray data compile the expression levels of various genes over a set of biological samples, for example, different drug treatments or normal vs. cancer cell lines. To observe physiological or pathological procession, we can also measure gene expression over a series of time points. Important applications of microarray data include classification and prediction of various human diseases, clustering of gene patterns and regulatory mechanisms, selection of identified biomarkers, reconstruction of gene regulatory networks (GRNs). However, there are quite a few problems in microarray data analysis that challenge bioinformatics scientists, for example, data noise, missing value, high false positive rate, measurement uncertainty, data imprecision, high dimensionality, difficulty of mining temporal information, low accuracy of current GRN models, and expensive computational cost. We believe that Computational intelligence (CI) can effectively address these challenging issues. We propose to tackle these problems using the following methods, but not limited to:(1) Neural networks and kernel based approaches can be used for classification, clustering, and gene selection; (2) Genetic and swarm intelligence algorithms can be used to search a discriminative subset of genes; (3) Modeling optimal GRNs is a NP-hard problem, and hence CI could be employed as alternative approaches to search good structures, given appropriate representations of the (dynamic) networks. This special session is soliciting high-quality papers of original research and application papers that have not been published elsewhere and are not under consideration for publication elsewhere. All papers will be rigorously reviewed by at least 3 reviewers. Accepted papers will be published in the CIBCB 2012 proceedings (with ISBN number), included in the IEEE Xplore digital library, and indexed by EI/Compendex. This special session is of clear interest to the computational intelligence community, the biology communities, as well as the multilinear (tensor) algebra community.

    The topics of this special session include, but are not limited to:

    • - microarray time-series data analysis
    • - microarray DNA methylation data analysis
    • - clustering, biclusering, and triclustering of gene expression profiles
    • - clinical diagnosis and prognosis
    • - gene selection
    • - pathway analysis
    • - microarray visualization, image processing and data preprocessing
    • - modeling and reconstructing gene regulatory networks
    • - network based systems biology

    • - Dr. Yifeng Li
      School of Computer Science
      University of Windsor
      Windsor, ON, N9B 3P4, Canada

    • - Dr. Chengpeng (Charlie) Bi
      Division of Clinical Pharmacology
      The Children's Mercy Hospitals and Clinics
      Kansas City, MO 64108, USA

    • - Dr. Sung-Bae Cho
      Computer Science Department
      Yonsei University
      Seoul, 120-749, Korea
      Email: sbcho@cs.yonsei.ac.kr

    • - Dr. Kyung-Joong Kim
      Department of Computer Engineering,
      Sejong University
      Seoul, 143-747, Korea

    • - Dr. Alioune Ngom
      School of Computer Science
      University of Windsor
      Windsor, ON, N9B 3P4, Canada


    Scope and motivation:
    Recent advances in gene sequencing technology are shedding light on the complex interplay between genes that elicit phenotypic behavior characteristic of any given organism. It is now known that in order to mediate external as well as internal signals, an organism’s genes are organized into complex signaling pathways that interact to regulate development and respond to environmental cues. Gene regulatory networks are structures that represent these gene-gene interactions occurring within an organism’s cell. They have been modeled variously as Boolean networks, probabilistic networks, and differential equation models. Due to the complex nature of gene regulatory networks, gaining further insights into them is a daunting task, requiring much interdisciplinary effort. Recent years have begun to see a substantial amount of research activity at the confluence of computer science, mathematics and biology, aimed at modeling gene regulatory networks.

    This special session will focus on current research aimed at the structure discovery, parameter estimation and systems identification of gene regulatory networks. Topics include, but are not limited to:

    • - soft computing
    • - evolutionary algorithms
    • - machine learning
    • - neural networks
    • - systems theory
    • - stochastic and deterministic optimization
    • - detection & estimation theory
    • - pattern recognition
    • - mathematical programming
    • - statistical methods
    • - information theory and other approaches that have been undertaken to infer, understand and apply gene regulatory networks.

    • - Sanjoy Das
      Associate professor
      Electrical & Computer Engineering Department,
      Kansas State University.
      Manhattan, KS 66506 USA

    • - Doina Caragea
      Associate professor
      Department of Computing and Information Sciences,
      Kansas State University

    • - Stephen M. Welch
      Email: welchsm@ksu.edu

Please submit special session proposals, which should contain a title, motivation and scope as well as a biosktech of the organizers, to Dr Jonathan Chan by email (Email: jonathan@sit.kmutt.ac.th).