Titel Untertitel
Total Page:16
File Type:pdf, Size:1020Kb
KNIME Image Processing Nycomed Chair for Bioinformatics and Information Mining Department of Computer and Information Science Konstanz University, Germany Why Image Processing with KNIME? KNIME UGM 2013 2 The “Zoo” of Image Processing Tools Development Processing UI Handling ImgLib2 ImageJ OMERO OpenCV ImageJ2 BioFormats MatLab Fiji … NumPy CellProfiler VTK Ilastik VIGRA CellCognition … Icy Photoshop … = Single, individual, case specific, incompatible solutions KNIME UGM 2013 3 The “Zoo” of Image Processing Tools Development Processing UI Handling ImgLib2 ImageJ OMERO OpenCV ImageJ2 BioFormats MatLab Fiji … NumPy CellProfiler VTK Ilastik VIGRA CellCognition … Icy Photoshop … → Integration! KNIME UGM 2013 4 KNIME as integration platform KNIME UGM 2013 5 Integration: What and How? KNIME UGM 2013 6 Integration ImgLib2 • Developed at MPI-CBG Dresden • Generic framework for data (image) processing algoritms and data-structures • Generic design of algorithms for n-dimensional images and labelings • http://fiji.sc/wiki/index.php/ImgLib2 → KNIME: used as image representation (within the data cells); basis for algorithms KNIME UGM 2013 7 Integration ImageJ/Fiji • Popular, highly interactive image processing tool • Huge base of available plugins • Fiji: Extension of ImageJ1 with plugin-update mechanism and plugins • http://rsb.info.nih.gov/ij/ & http://fiji.sc/ → KNIME: ImageJ Macro Node KNIME UGM 2013 8 Integration ImageJ2 • Next-generation version of ImageJ • Complete re-design of ImageJ while maintaining backwards compatibility • Based on ImgLib2 • http://developer.imagej.net/ → KNIME: Tight integration – Automatic node generation from ImageJ2-plugins KNIME UGM 2013 9 Integration BioFormats • Library for reading and writing > 120 image file formats • OME-XML standard • http://loci.wisc.edu/software/bio-formats → KNIME: Image Reader and Writer KNIME UGM 2013 10 Integration OME / OMERO • Tools for storing (OMERO database), visualizing, managing and annotating microscopic images and metadata • http://www.openmicroscopy.org/ → KNIME: Experimental OMERO Reader KNIME UGM 2013 11 The KNIME Image Processing Extension – An Example KNIME UGM 2013 12 Some Nodes Input/Output Image Proc. Segmentation Features Views KNIME UGM 2013 13 Example: High-Content Screening KNIME UGM 2013 14 Example: High-Content Screening positive negative KNIME UGM 2013 15 Example: High-Content Screening KNIME UGM 2013 16 Example: High-Content Screening KNIME UGM 2013 17 Example: High-Content Screening KNIME UGM 2013 18 Example: High-Content Screening KNIME UGM 2013 19 Example: High-Content Screening KNIME UGM 2013 20 Example: High-Content Screening KNIME UGM 2013 21 Example: High-Content Screening KNIME UGM 2013 22 … KNIME UGM 2013 23 Resources • Workshop tomorrow! • Homepage: http://tech.knime.org/community/image- processing • Forum: http://tech.knime.org/forum/knime-image- processing • Contact: [email protected] [email protected] [email protected] KNIME UGM 2013 24 Thank you… KNIME UGM 2013 25 .