1. HOME
  2. Creating numerical human body phantoms for use with X-ray CT scans

Creating numerical human body phantoms for use with X-ray CT scans

  • School of Medicine/Graduate School of Medicine
  • Radiological and Medical Laboratory Sciences
  • Radiological Sciences

Mitsuru Ikeda [Professor]

Outline of Seeds

We will have established a technique for creating accurate reconstruction image simulations with virtual X-ray CT imaging device using numerical phantoms configured from (energy-separate) X-ray attenuation coefficients that are accurately corresponding to reconstruction images of the human body taken with actual X-ray CT imaging devices. Through a trial-and-error process rooted in macroanatomical knowledge, we are planning to use this technique to create numerical phantoms configured from (energy-separate) X-ray attenuation coefficients corresponding to a standard healthy human body and create simulations with these numerical phantoms in order to try to accurately replicate X-ray CT images. Next, again using a trial-and-error process based on macropathological knowledge, we are also planning to create similar numerical phantoms corresponding to disease and try to replicate their X-ray CT images through simulations. For creating numerical phantoms corresponding to both healthy human bodies and disease models, we will begin with targets corresponding to macropathology and anatomical analysis at a voxel size visible to the naked eye and will continue next with microscopic targets. Through this process, we shall gain new insights concerning CT image diagnosis that will not be available to us using methods based on conventional comparisons of CT images with pathology images. We shall also gain new insights into the relationship between signals and noise in X-ray CT images, which will allow us to rationally reduce radiation exposure during X-ray CT imaging.

Novelty and originality of this research

The goal of our research creating numerical phantoms configured from (energy-separate) X-ray attenuation coefficients, corresponding to both standard healthy human bodies and disease models, is something that has never been done before. This is also the first time that anyone has tried to create accurate reconstruction images of the human body, including healthy bodies, with simulations that use virtual X-ray CT imaging devices from these phantoms. The ability to create numerical phantoms configured from (energy-separate) X-ray attenuation coefficients as well as use them to create accurate X-ray CT image simulations is expected to yield several benefits: (1) advancing X-ray CT diagnostics, (2) contributing to the development of systems that support X-ray CT image diagnoses, (3) helping to explain the relationship between signals and noise in X-ray CT imaging, (4) Monte Carlo simulations that allow accurate evaluations of radiation exposure during X-ray CT imaging, and (5) a significant contribution to rationally lessening radiation exposure through an understanding of the noise generated during X-ray CT imaging.

Application and research area for Industry collaboration

Our research requires the cooperation of medical device manufacturers who create commercially available X-ray CT imaging devices, as we are requesting that they provide sinogram data from machines that are on the market. We will use general reconstruct functions through simulations that use virtual X-ray CT imaging devices accurately corresponding to commercial equipment. Applying the findings of our research is expected to allow for (1) the development of rational and efficient means for reducing radiation exposure, (2) dual energy imaging applications, (3) the development of effective noise reduction methods, and more. And once our research is further developed, it will be applicable to the creation of numerical phantoms for MR imaging as well.

Key Takeaway

The goal of our research is to create numerical phantoms configured from (energy-separate) X-ray attenuation coefficients, corresponding to both standard healthy human bodies and disease models, phantoms that will allow us to use simulations to accurately replicate X-ray CT images produced using commercially available equipment. The phantoms will be extremely useful in a broad variety of settings, from basic technologies to clinical applications.

Keywords

Numerical phantoms, radiation exposure, CT image simulation, nose evaluation, noise reduction

Technologies

  • Simulations generating sinograms from numerical phantoms using virtual X-ray imaging equipment
  • Simulations generating X-ray CT images using sinograms from virtual X-ray imaging equipment

Monographs, Papers and Articles

  • A new evaluation method for image noise reduction and usefulness of the spatially adaptive wavelet thresholding method for CT images, Australasian Physical and Engineering Sciences in Medicine, 35: 475-483, 2012.