PhD-level summer school sponsored by the Swedish Society for Automated Image Analysis and SciLifeLab Uppsala

How to use and make additions to open-source software for image processing, analysis, and visualization in medicine and microscopy.

June 3-5, 2013 at the Centre for Image Analysis, Division of Visual Information and Interaction, Dept. of Information Technology, Uppsala University.

While developing methods as part of your thesis work it is common to write code in Matlab, C and its variants. Once results are obtained and published, it is unfortunately also very common to hop on the next project (or job), and leave the code as it is, often in a shape difficult for others to use, and/or restricted by licensing fees. There is however a large community of algorithm developers who contribute their code to free and open-source projects. These projects are also great sources for already implemented open source algorithms, making them a great starting point for further exploration and development, as well as reproducible comparison of competing methods. Many journals also require access to code at publication, making opens source initiatives a good alternative. By becoming more familiar with these tools, and learning how to make your own contributions/modifications, you will have a valuable toolbox that can be shared and used also after leaving the university.

During June 3-5, 2013, Felix Ritter, Anja Hennemuth, Lee Kamentsky, and Kevin Eliceiri will visit the Centre for Image Analysis at Vi2, Uppsala University. During these three days, they will provide lectures and hands-on tutorials on MeVisLab, ImageJ/FIJI and CellProfiler. The course participants will get an overview of the software and learn the basics of how to contribute their own code to these open-source initiatives. The programming languages will be Java and Python.

About teachers and software:



Anja Hennemuth and Felix Ritter, MeVisLab

MeVisLab represents a modular framework for the development of image processing algorithms, visualization, and interaction methods with a special focus on medical imaging. Besides basic image processing and visualization modules, MeVisLab includes advanced medical imaging algorithms for segmentation, registration, and quantitative morphological and functional image analysis. MeVisLab is being developed and used by MeVis Medical Solutions AG and Fraunhofer MEVIS in Bremen, Germany. MeVis Medical Solutions and Fraunhofer MEVIS develop scientific methods and software for computer assistance in medicine in general and radiology in particular, including computer aided diagnosis, therapy planning and monitoring, and computer aided teaching and training.

Anja Hennemuth is employed at Fraunhofer MEVIS since 2009 and holds a Ph.D. from the University of Bremen in computer science. She leads the field of cardiovascular research and development and has advised both research projects as well as projects in cooperation with industrial partners using the MeVisLab platform.

Felix Ritter is head of visualization and human computer interaction at Fraunhofer MEVIS, Bremen, Germany. His research interests include medical visualization, perception, and human factors in visualization. Felix’s current work is focused on the usability of medical workstations and the prototyping process using MeVisLab.

Anja Hennemuth and Felix Ritter will teach on June 3.

Lee Kamentsky, CellProfiler

With nearly 30 years of professional software engineering experience, including writing the image-analysis and hardware-control software for CompuCyte's Laser Scanning Cytometer, Lee Kamentsky joined the Imaging Platform of the Broad Institute of Harvard and MIT in 2008, where he is the lead software developer for the CellProfiler project, implementing and optimizing digital image algorithms, ensuring documentation and unit testing. CellProfiler is free open-source software designed to enable biologists without training in computer vision or programming to quantitatively measure phenotypes from thousands of images automatically. CellProfiler has been cited more than 600 times since its launch in 2006, is downloaded more than 25 times per day, and launched 250 times per day (excluding launches at the Broad Institut). Its open source written in Python, communicates with ImageJ, and is available for Windows, Mac and Linux. During his tutorial, Lee will start with a quick overview of the CellProfiler software followed by an introduction to programming in Python and using SciPy. He will thereafter present the organization of the CellProfiler software and how modules are structured.

Lee Kamentsky will teach on June 4.

Kevin Eliceiri, FIJI/ImageJ2

Kevin Eliceiri is director of the Laboratory for Optical and Computational Instrumentation (LOCI) and principal investigator in the Laboratory of Molecular Biology at the University of Wisconsin, where his current research focuses on the development of novel optical imaging methods and software for multidimensional image analysis. He is also one of the principal investigators at ImageJDev, is a multi-institution project to develop the next-generation version of ImageJ, an image processing program widely used for scientific research. The aim is to strengthen both the ImageJ software itself and its community by pursuing a unified vision of ImageJ and associated community resources, including code and plugin repositories, and user and developer documentation. ImageJ was originally called NIH Image, launch 1987. It is Java-based and extensible via Java plugins, http://rsb.info.nih.gov/ij/ was visited > 9M times, and has 7000 visitors/day. More than 210,000 published papers cite ImageJ. During his tutorial Kecin Eliceiri will present an overview of what FIJI/ImageJ2 can do, and show basics in script writing and plugin writing.

Kevin Eliceiri will teach on June 5.





Course organizer:
Carolina Wählby, Associate professor and senior lecturer, the Centre for Image Analysis and SciLifeLab Uppsala. The number of participants will be limited to 20. A mailing list with the names of the participants will be shared with the teachers for communication of course details. Thanks to sponsoring from the Swedish Society for Automated Image Analysis we will be able to keep the course fee down to 500 SEK per participant. Lunches (June 3-5) and dinners (June 3 and 4) will be included in the cost. Participants will have to organize housing on their own. To sign up for the course (or ask questions) please send an email to carolina@cb.uu.se, including name of course participant and adress (+reference person) for invoice of course fee. You can also call +46-18-4713473.