Unexpected error when reconstructing CT image

Dear CASToR users,

I would like to reconstruct an image simulated using GATE. I’ve defined a detector, a fan-beam X-ray source and a simple phantom (water disc with a bone disc inside). The phantom is rotating around its axis so I can get a CT image. I’ve given CASToR the cdh and cdf file (cdf and geom file attached). I have two questions:

  1. Somehow CASToR initialises the image dimensions as (180,180,240), but they should be (240,180,0). What am I doing wrong?
  2. At the end it says “Unknown event mode”. I assume this is because of my cdf file? I wasn’t sure how to format it exactly. It’s too big to attach, but I am posintg the first few lines below. First column is the time, second is projection ID and the third is pixel ID and each line corresponds to one event.
    0 2 31563
    0 2 21542
    0 2 21680
    0 2 21407
    0 2 21721
    0 2 21630
    0 2 21450
    0 2 21833
    0 2 21645
    0 2 21428
    0 2 7364

Has anyone an idea what I need to change?

Many thanks in advance!

Best wishes,

The Institute of Cancer Research: Royal Cancer Hospital, a charitable Company Limited by Guarantee, Registered in England under Company No. 534147 with its Registered Office at 123 Old Brompton Road, London SW7 3RP.

This e-mail message is confidential and for use by the addressee only. If the message is received by anyone other than the addressee, please return the message to the sender by replying to it and then delete the message from your computer and network.

CT_PHANTOM.geom (388 Bytes)

CT_PHANTOM.cdh (242 Bytes)

error (5.05 KB)

Dear Margarete,

CASToR does not handle list-mode CT data for now as there is no algorithm able to deal with such data. The “Unknown event mode” is coming from that.

You will have to compute projections for each angle position of your phantom/scanner and then build a CT histogram datafile. You need to simulate a blank scan as well.

The CDF file is in a binary format, as opposed to the ascii format.

All mandatory and optional fields for each histogram bin (called generic event in castor) of the datafile are detailed in the documentation.



Dear Simon,

Thanks for your reply. I am not sure how to produce the histogram and am a bit confused about what each histogram bin contains. Is it the number of events along the width of the detector (so the pixels being the bins)? For my analysis, I reconstruct the individual projection images and based on these, I build the sinogram, but I am not sure how to retrieve the histogram.
Also, could you please explain why I need a blank scan?

Many thanks,

Hi Maggie,

The histogram is your sinogram but in the CASToR format. One histogram bin contains the number of detected counts in a given sinogram bin which is defined by a projection ID and a pixel ID.

The blank scan calibrate your couple source/detector.

You do a simulation without anything in the field of view for a single projection. You need that to know what is the expected number of counts without object.



Hi Simon,

Thanks for clearing that up. I understand that one sinogram bin is defined by a unique projection ID, however I am not sure about the pixel ID. In my case, all the data in one sinogram bin is acquired along my detector width and therefore spread across 240 pixels. So if each pixel has a unique ID, this means that one projection will have several pixel IDs associated with it, right?

Many thanks,