No announcement yet.

Image Analysis Results

This topic is closed.
  • Filter
  • Time
  • Show
Clear All
new posts

  • Image Analysis Results

    Here are the responses I recieved regarding CT image analysis. I have
    included the original and the two responses. I would like to thank those
    who responded.
    Brian Szmal, Department of Bioengineering-Syracuse University

    ****Original Message ******************************************

    Hello. We are trying to determine bone bending stiffness and the moment of
    inertia from CT image data using NIH image. Can Slice_CT (Corcoran et al.,
    1994) be used in this way? Others have performed the analysis from the
    data acquired from the CT scan, but not within a computer program such as
    NIH image. We think that NIH image can be used in this type of analysis,
    but we currently do not have the required macros to perform the functions.
    We were wondering whether or not anyone else has used NIH image or other
    imaging analysis programs such as Adobe, Osiris, Analyze, or even a
    modified version of Slice that can be performed on a Mac. Your assistance
    in this matter will be appreciated. I will be glad to post the results
    that are received.
    Thank you in advance.
    Brian Szmal
    Department of Bioengineering, Syracuse University

    ************************************************** ***********
    *****from "John A. Hipp"********************

    I have several versions of C programs that work by themselves without any
    additional image processing program. The different versions of the code
    are for specific QCT phantoms. For example, one is for dipotasium
    phosphate phantoms and the other is for hydroxyapatite phantoms. The
    programs take raw binary files as input. The raw binary files can be
    created by stripping off the header from the manufacturers
    format and uncompressing the data if needed. These programs use simple
    parameter files to get the image size, pixel size, CT attenuation for
    each of the QCT phantom chambers, etc. These programs calculate much more
    then simple cross sectional areas and moments of inertia. They also
    calculate the density-weighted versions of these parameters: axial
    rigidity, and moments of inertia. They also calculate the density-weighted
    versions of these parameters: axial rigidity, bending rigidities,
    principal bending rigidities. I have used these programs extensively in
    the UNIX environment where loops can be used to automatically analyze a
    series of CT slices. Auxillary programs are used to determine linear
    regressions between CT attenuation and bone density that are unique to
    each slice. We found that a slice specific calibration is needed for the
    most accurate results. I can send you some of this code if it would help.
    I also have versions of the code that work as modules in the AVS 5 image
    processing software from Advanced Visualization Systems. AVS is a
    fantastic program, but it is expensive and only works on UNIX machines.
    We have a paper coming out in Clinical Orthopaedics and Related
    Research where this software was used to show that CT-based axial
    rigidity measurements for the centrum of the vertebral body predicts
    almost 90% of the variation in failure load for intact vertebrae and
    vertebrae with simulated and actual metastatic defects.

    John Hipp, PhD

    ************************************************** ******
    *****from "Jeff Linn"

    You are probably aware that NIH Image has an lively mailing list filled
    with resources for all varieties of imaging problems or challenges. There
    are a number of participants who would be able to assist you in the
    creation or modification of macros for your task. I have had wonderful
    assistance from the list and highly recommend it. If you haven't yet
    checked out the ftp site at ftp://zippy.nimh.nih.
    gov/pub/nih-image then do. There is an archive of macros. There are also
    a number of companies listed on the NIH Image homepage which do commercial
    source code and macro development. Being only passingly familiar with 3D
    reconstruction (I play with the visible human dataset) I've found the NIH
    Image spin off 'Object Image 1.60" very useful. It allows you to view
    volumes in sagittal, coronal and axial, plot points and vectors and do
    calculations on the resulting co-ordinates. It is available zippy, I
    believe in the spinoffs directory.

    To unsubscribe send UNSUBSCRIBE BIOMCH-L to
    For more information: