Gel phantom data for dynamic X-ray tomography

<p>The gel phantom was constructed to simulate diffusion of liquids inside plant stems, namely the flow of iodine-based contrast agents used in high resolution tomographic X-ray imaging of plants. In order to test different reconstruction methods, this radiation resistant phantom with similar diffusion properties was constructed.</p> <p>A more detailed documentation can be found on arXiv: <a href="https://arxiv.org/abs/2003.02841">https://arxiv.org/abs/2003.02841</a></p> <p>The phantom consists of a 50 ml Falcon test tube filled with agarose gel. After the agarose solidified, five cavities were made into the gel and filled with 20% sucrose solution to guarantee the diffusion by directing osmosis to the gel body. In addition densely punctured plastic straws were placed in the cavities to simulate cellular passages such as <em>phloem plasmodesmata</em> and to slow down the lateral diffusion.</p> <p>The primary measurements consisted of 17 consecutive time frames, with initial stage of no contrast agent followed<br> by steady increase and diffusion into the gel body over time. Each round of measurements consist of 360 projections with a fanbeam microCT-scanner, but we used only the central plane of the cone beam, resulting in 2D fan beam geometry.</p> <p>Data is given in two different resolutions corresponding to reconstructions of size 256 x 256 or 512 x 512, (<em>GelPhantomData_b4.mat</em> and <em>GelPhantomData_b2.mat</em> respectively). In addition to the primary measurements, a more densely sampled measurements from the first time step and an additional 18th time step are provided in <em>GelPhantom_extra_frames.mat</em>.</p> <p>The measurements are stored in special data structures containing all the necessary metadata. In combination with the MATLAB toolboxes this allows for easy application of forward operators and reconstruction algorithms.</p> <p>This is demonstrated using the included example codes. These require ASTRA Toolbox, Spot Linear Operator Toolbox and HelTomo Toolbox (v1) for MATLAB.</p> <p><strong>NOTE</strong>: Some of the metadata field names are different in HelTomo v2 and higher. Use of the data requires renaming (or adding) two fields in to the parameters:<br> numDetectorsPost = numDetectors;<br> effectivePixelSizePost = effectivePixelSize;<br> <strong>Update: </strong>in v1.2 the data and example codes have been updated to match HelTomo v2.1 convention. For backwards compatibility (with Heltomo 1.0) the old medatadata field names remain.</p> <p>The full cone-beam measurements and corresponding documentation are available in <a href="https://doi.org/10.5281/zenodo.7739984">Zenodo</a>.</p>