This course provides attendees with key steps involved in commercializing a medical imaging innovation. It offers fundamental concepts in life cycle management of innovative ideas, translation and validation in clinical trials, financing and strategic partnership. It also covers topic related to securing and licensing of intellectual property (IP), formation of business entity, raising capital, regulatory pathways and market analysis. This course includes specific case studies highlighting key traces for success and failure. Additionally, specific tools for technology valuation for in/out licensing, merger/acquisition (M&A) and other transactions will be discussed. Specific areas of focus will be image-based algorithms/software, imaging modalities and instrumentations for diagnostics and operative applications.

This course gives an overview of medical image formation, enhancement, analysis, visualization, and communication with many examples from medical applications. It starts with a brief introduction to medical imaging modalities and acquisition systems. Basic approaches to display one-, two-, and three-dimensional (3D) biomedical data are introduced. As a focus, image enhancement techniques, segmentation, texture analysis and their application in diagnostic imaging will be discussed. To complete this overview, storage, retrieval, and communication of medical images are also introduced. In addition to this theoretical background, a 45 min practical demonstration with ImageJ is given. ImageJ is a Java-based platform for medical image enhancement and visualization. It is developed by the National Institutes of Health, USA, open source and freely available in the public domain. For this course, ImageJ is appropriately configured with useful plug-ins (e.g. DICOM import, 3D rendering) and distributed on CD-ROM. Attendees are welcome to perform on their own laptop computers.

This course is an overview of texture; its appearance, modeling and applications in diagnostic imaging. It begins with general examples for defining and modeling texture. Approaches for identifying and measuring image textures are described. These approaches include statistical methods, Markov random field and morphological techniques. Applications of these techniques in two specific problems with diagnostic imaging will be discussed. A model to synthetically generate digital mammographs followed by multi-resolution analysis and segmentation of images is discussed. Then modeling and textural analysis of hard tissues (bone) as a screening tool for osteoporosis is examined. Histomorphometric approaches for both thin slice and 3-D imaging will be discussed.