The Fifteenth Computational Anatomy Seminar



  • Lecture 1
    • Speaker
      • Kenji Suzuki, Ph.D
        Assistant Professor of Radiology, Medical Physics, and Comprehensive Cancer Center
        Department of Radiology, University of Chicago
  • Title
    • Pixel-based Machine Learning in Medical Imaging and Computer-aided Diagnosis
  • Abstract
    • Machine leaning (ML) plays a very important role in medical imaging and computer-aided diagnosis. Recently, pixel/voxel-based ML (PML), which uses pixel/voxel values in images directly instead of features calculated from segmented regions as input information, emerged. Because the PML can avoid errors caused by inaccurate feature calculation and segmentation, which may be often occur for subtle, complicated patterns, the performance of the PML can potentially be higher than that of common classifiers. In this talk, PMLs are introduced and reviewed to make clear a) the basic principles of the PML and b) its applications to 1) quantum noise reduction in x-ray images, 2) enhancement of subjective edges traced by a physician, 3) separation of bones from soft tissue in chest radiographs, 4) lung nodule detection on CT images, 5) distinction between benign and malignant nodules on CT images, and 6) polyp detection in CT colonography.
  • Biography
    • Kenji Suzuki, Ph.D. (by Published Work; Nagoya University) worked in Research and Development Center, Hitachi Medical Corporation, Japan, from 1993 to 1997. From 1997 to 2001, he worked in Faculty of Information Science and Technology, Aichi Prefectural University, Japan, as a faculty member. In 2001, he joined Department of Radiology, the University of Chicago, as Research Associate. Since 2006, he has been Assistant Professor of Radiology, Medical Physics, and Cancer Research Center there. His research interests include computer-aided diagnosis, machine learning in medical imaging, and medical image analysis. He published more than 230 papers (including 92 peer-reviewed journal papers). He is inventor on 30 patents (including 11 granted patents), which were licensed to several companies and commercialized. He published 9 books and 18 book chapters, and edited 6 journal special issues. He was awarded/co-awarded more than 44 grants including NIH R01 grants totaling 10.2 million dollars. He has served as the Editor-in-Chief/Associate Editor of 26 leading international journals, including Medical Physics, Academic Radiology, and International Journal of Biomedical Imaging. He has served as a referee for 62 international journals, an organizer of 16 international conferences, and a program committee member of 107 international conferences. He received 32 awards for his research and teaching, including 3 Certificate of Merit Awards from RSNA, IEEE Outstanding Member Award from IEEE, and Kurt Rossmann Award for Excellence in Teaching from University of Chicago.

  • Lecture 2
    • Speaker
      • Miguel A. González Ballester, DPhil (Oxon.)
        Chief Research Officer
        Alma IT Systems
    • Title
      • Personalised surgical planning and biomechanical simulation based on computational statistical models
    • Abstract
      • This talk will present an overview of some of the research activities of a research-intensive company based in Barcelona, Spain, which develops and commercialises software for medical image visualization and processing, with applications in image-based diagnosis and surgical planning. Its research department focuses on novel methods for image analysis, computational simulation and surgical navigation. Example projects will be presented, such as a system for planning of tracheal implant surgeries including biomechanical simulation of swallowing movements, a statistical finite element model for orthopaedic biomechanics, and a novel framework for image-based optimisation of cochlear implant surgeries. General discussion will include aspects related to the interaction between academic and industry-driven research.
  • Biography
    • Dr. Miguel A. González Ballester holds a doctoral degree from the University of Oxford, UK (2000). His doctorate, under supervision of Sir Michael Brady and Prof. Andrew Zisserman, focused on the analysis of brain MRI data for multiple sclerosis and schizophrenia. He was awarded the prestigious Toshiba Research Fellowship and moved to Japan to work for two years as a senior researcher at Toshiba Medical Systems (Otawara, Tochigi), where he developed novel, patented systems for MRI parallel imaging. In late 2001 he obtained a faculty position at INRIA (Sophia Antipolis, France), where he led research projects on medical image analysis and mathematical modelling. In 2004 he joined the University of Bern (Switzerland), as head of the medical image analysis group, and later became head of the surgical technology division at the Faculty of Medicine. There, he supervised a division composed of 4 research groups (50+ staff) working on medical image analysis, computer-assisted surgery (mainly for orthopaedics and CMF surgery), and surgical robotics and mechatronics. From 2008 he heads the Research Department at Alma IT Systems in Barcelona (Spain), where he leads the development of the new generation of computer tools for diagnosis and surgical planning. He is the coordinator of the European FP7 research project HEAR-EU, has approximately 150 publications in peer-reviewed scientific journals and conferences, and supervised 14 PhD theses. He is a regular contributor to scientific committees (such as the IEEE Technical Committee on Bio Signal and Image Processing) and review boards at numerous journals and conferences.


  • Yoshinobu Sato, PhD (Associate Professor, Department of Radiology, Osaka University)
    • Phone +81 6 6879 3562
    • E-mail yoshi @

Scenes at Computational Anatomy Seminar

  • photo_20130702-1.jpg
  • photo_20130702-2.jpg
  • photo_20130702-3.jpg

添付ファイル: filemiguel.jpg 1315件 [詳細] fileSuzukiK_UChicago.jpg 1252件 [詳細]

トップ   編集 差分 バックアップ 添付 複製 名前変更 リロード   新規 一覧 単語検索 最終更新   ヘルプ   最終更新のRSS
Last-modified: 2013-09-03 (Tue) 11:24:15 (1507d)