ISBI 2016 Special Session on Big Data in Medical Imaging

The ISBI 2016 Special Session on Big Data in Medical Imaging will be held on Wednesday 13th April 2016, at the Clarion Congress Hotel in Prague.  There will be an exciting program, where senior researchers from academia and industry will be exchanging ideas on the role of Big Data in Medical Imaging.  

Recent years have seen an increasing volume of medical image data and annotations being collected and stored. Recent initiatives such as the Cardiac Atlas Project [1] and the (developing) Human Connectome Projects [2,3] have begun the acquisition of hundreds of terabytes of data to be made available to the scientific community.

Processing these large datasets is key to providing a wealth of information with the potential to be usefully harnessed [4]. Along with the new clinical opportunities arising, novel image and data processing algorithms are required for working with, and learning from, large scale datasets. The community expects new tools and methods that not only allow sensible analysis on the single subject level, but which are also robust and scalable with respect to groupwise studies on a very large database. Additionally, international data sharing, ethical and ownership issues need to be considered.

This Special Session aims to examine recent progress in the field, together with new openings stemming from increased data availability, as well as the specific challenges involved. To discuss these ideas, we have exciting speakers leading the Big Data drive in medical imaging. Their medical imaging and computer vision expertise is supplemented by particular experience in Big Data topics including multicentre clinical trials, large scale image search, and connectomics.


William M. Wells III: Invariant Feature-Based Analysis of Medical Images: An Overview

Professor of Radiology at Harvard Medical School, he also maintains strong links to the MIT Computer Science and Artificial Intelligence Laboratory. His work has focused on analysis of structural and functional MRI and image-guided therapy. His more recent work investigates the use of image features for organising and deriving information from large scale databases.

• Flora Gilboa-Solomon: Cognitive Radiology Assistant

Manager of Medical Imaging Analytics at IBM Research, Haifa. She leads the team developing the Medical Sieve - the next generation of cognitive assistant for radiologists - in particular to address the challenge of breast cancer. The work develops advanced decision support tools that combine multi-modal image analytics and clinical data analysis. Technically, this includes advanced computer vision techniques to allow automatic extraction of diagnostic relevant features, as well as machine learning tools used to combine multimodal semantic image descriptions with clinical data, thus facilitating estimation of correct differential diagnosis and patient management recommendation.

Tal Arbel: Hierarchical Probabilistic Graphical Models for the Detection and Segmentation of Multiple Sclerosis Lesions in Multi-centre Clinical Trial Datasets

Associate Professor and Research Director of the Probabilistic Vision Group "Medical Imaging Lab” at McGill University. She has strong ties to the computer vision community and aims at developing modern probabilistic techniques in computer vision and apply them to problems in medical imaging. She will discuss the challenges associated with the development of robust and efficient methods for image analysis in the context of clinical trials.

• Georg Langs: Predictive Signatures from Real World Clinical Imaging Data

Associate Professor and the Head of the Computational Image Analysis and Radiology Lab (CIR) at the Medical University of Vienna. He is also affiliated with the Medical Vision Group at CSAIL, Massachusetts Institute of Technology. His research interests include machine learning, and computer vision to learn from large image sets, to structure visual information, and to make it usable during the diagnosis process. His current work focuses on neuroimaging and the study of multi-variate functional activity that emerge during specific cognitive processes.

Organising committee

  • Kanwal Bhatia
    Postdoctoral Research Associate, Department of Biomedical Engineering, King's College London

  • Herve Lombaert
    Research Scientist, INRIA Sophia Antipolis

  • Sarah Parisot
    Postdoctoral Research Associate, Biomedical Image Analysis Group, Imperial College London

  • Jonathan Passerat-Palmbach
    Postdoctoral Research Associate, Biomedical Image Analysis Group, Imperial College London

[4] Craddock et al. GigaScience (2015) 4:13 DOI 10.1186/s13742-015-0045-x