Bio-medical Imaging in Java Central

Only one download for VolumeJ, FlowJ, PCA, MI and minor plugins.

Biomedical imaging software, manuals, documentation and presentations. Most are implemented using ImageJ, the image processing program from NIH written by Wayne Rasband.

Downloading
All plugins and supporting libraries are in one file, bij.jar. The only thing you need to access themis bij.jar and the plugin stub files (otherwise ImageJ cannot find them) in the plugins directory.

• BIJ plus ImageJ 1.32 as a self-extracting Windows .exe file (8Mb). For Windows users only; please install in c:\ to avoid configuring anything.
• BIJ plus ImageJ 1.32 as a single package (8Mb), for all other operating systems.
• BIJ plugin only (300Kb) as a ZIP file, if you already have ImageJ v1.32 or higher.
• BIJ sources (263Kb).
• The Javadoc API for BIJ. For those who want to use BIJ routines in their own programs.
  By downloading, you indicate your agreement to the copyright and disclaimer text.

The following image processing plugins are available automatically within bij.jar:

VolumeJ - volume rendering in Java. Manuals, FAQ, examples, references. NEW test version of interactive rendering using trackball type interface.

SurfaceJ - 3-D rendering of images and image sequences as surfaces ('mountain scapes'). Manuals, FAQ, examples, references.

FlowJ - optical flow estimation and visualization software with several popular (2-D) optical flow algorithms. Manuals, FAQ, examples, references.

Flow3J - a new three-dimensional optical flow algorithm, based in part on the VJ API.

PCA for Principal Component Analysis of image stacks, gives the eigenimages and eigencoordinates.

Register ROI and Register_ROI MI Edges for alignment of all slices in a stack to the first slice. The MI stands for Mutual Information, the best performing registration metric, and Edges stands for the use of edges - largest Hessian eigenvalue - instead of gray values for improved performance. Implements affine registration, i.e. only translation , rotation and scaling. It determines the registration parameters from registration of the ROI, and then aligns the full slice, so that for large images, you can define a small ROI (in an area where there is a characteristic non-alignment or motion to correct for) making it much faster.

Other plugins have been implemented, but manuals are not available. There is also some documentation available on the ImageJ mailing list.

Other stuff

IMAQ API. Windows only. To acquire video images or video streams using NI IMAQ connected digital cameras in ImageJ. See enclosed README file.

UICAM.EXE Allows you to control 'any' camera on a Windows platform remotely. Currently only PVCAM cameras (Roper Scientific) supported. A great way to use all those different cameras : direct interfacing with ImageJ (JNI) is difficult, but this server allows a standalone (lowcost) camera PC that you can use as a box, no threat of updates to your regular system / software of intruding on your camera acquisition, remote control of camera for long experiments. I would like to set this up as an extension to ImageJ.

Run as an applet: here you can use all BIJ plugins including VolumeJ, Surface and FlowJ without downloading and installing. Allows opening and saving your own files. Also includes isosurface rendering algorithm, which is about 100 times faster than previous algorithms and allows real-time rendering. Design and implementation of this new, very fast, real time volume rendering algorithm has been completed. See the applet.

Older material

Somewhat dated (2001), my thesis Objective Measurement of Motion in the Orbit is available here in PDF format (illustrations 72 dpi only). Its title is "". The purpose of my thesis was to discover new techniques of measuring motion in soft tissues, and determine their clinical benefit. The research included mathematical and applied contributions to the measurement of optical flow in noisy and aliased Magnetic Resonance images, the measurement of optical flow in three-dimensional from 3-D image sequences, and from noisy and degraded 3-D images, to the visualization of two- and three-dimensional optical flow fields, and to image segmentation based on differential motion. It also included clinical applications to 2-D and 3-D motion estimation in orbital tissues, which lead to the discovery of the causes of both persistent pain after orbital surgery and persistent diplopia after orbital surgery, a new technique for non-invasive differentiation of benign and malignant orbital tumors, and to the description of the orbital soft tissues as an organ of gaze.

From my thesis, simulating motion of the human orbit for MRI studies using a sirloin steak in the rotation device. Figure © 2000 IEEE, by kind permission.

Michael Abràmoff, MD, PhD.
Assistant Professor of Ophthalmology / Electrical and Computer Engineering
Department of Ophthalmology and Visual Sciences
University of Iowa Hospitals and Clinics, Iowa
USA.