[Rtk-users] The Reconstruction Result has blur-image its possible due to motion blur effect or not? how can reduce it or correct it ?

Thu Feb 16 14:24:52 CET 2023

Hi Simon

1) thanks for information, I will study and look at the content of the file
link. and I think the main motion type t of dental CBCT is due to the up
and down swing caused by the acceleration of the gantry. I mainly want to
learn and understand the impact of this kind of motion effect on CBCT
reconstruction.

2) The slant is adjusted with the GantryAngle + source distance and offsets
since, in the detector coordinate system, the source is moving.
--> So if  I need to ad adjust the Slant Case ,I need to change variable
NO.3 "gantryAngle" in rtk::ThreeDCircularProjectionGeometry-> AddProjection
to "GantryAngle + source distance and offsets since",
if  the  source distance means  source to detector distance , and  the
source to detector distance is fixed,
that will be changed to
EX:  source to detector distance  = 500 / GantryAngle = 3 / offsets since =
5
GantryAngle will finally change to -->*3+500+5* ? am I correct?

3)thanks for your reply ,I will be try it

Thanks for reply again.

BR,

Simon Rit <simon.rit at creatis.insa-lyon.fr> 於 2023年2月16日 週四 下午8:21寫道：

> Hi,
> 1) I don't know what type of motion you are dealing with. You can read
> e.g. here <http://dx.doi.org/10.1118/1.3115691>
> <https://apis.ebsco.com/public/linkout/v1/ftf?ref=792a0735-69d6-405a-bd13-f5380020ce8b&id=302229>
> or here <http://doi.org/10.1088/1742-6596/489/1/012026>
> <https://doi.org/10.1088/1742-6596/489/1/012026> how we estimated motion
> in our case but I won't be able to help more.
> 2) The slant is adjusted with the GantryAngle + source distance and
> offsets since, in the detector coordinate system, the source is moving.
> 3) You will have to through the command lines.
> Simon
>
> On Thu, Feb 16, 2023 at 8:23 AM 何明哲 <m10512067 at yuntech.org.tw> wrote:
>
>>
>> hi Simon thanks for your reply, but I still have some question ,
>>
>> 1) This is for non-rigid motion. If the motion is rigid, it can be
>> compensated by adjusting the source and detector positions and orientation
>> wrt to the object in the geometry file.
>> --> So this example is not suitable for dental CBCT? Or the motion blur
>> of dental CBCT can be ignored in the RTK algorithm because the motion is
>> rigid?
>>
>
>> 2) Again, detector parallel to isocenter is meaning less, the isocenter
>> is a point. And the goal of the geometry description is no to "make them
>> parallel" but to account for the actual geometry of the system wrt scanned
>> object.
>> --> If I need to ad adjust the Slant Case (the pic below is )，What
>> methods or parameters can I use to adjust it?
>> [image: image200.png]
>>
>> (Pic form Wang, Zhao, et al. "Improvements in micro-CT method for
>> characterizing x-ray monocapillary optics." *Optics Communications* 504
>> (2022): 127474)
>>
>> 3) If I want to try to use dental CBCT or chest cavity CBCT to implement
>> and study the MCCBCTReconsttruction example
>> <https://wiki.openrtk.org/index.php/RTK/Examples/MCCBCTReconstruction#Deformation_vector_field>,
>> how can I collect or generate deformat.mha and motion signal.txt ?
>>
>> Thanks for reply again.
>>
>> BR,
>>
>>
>> Simon Rit <simon.rit at creatis.insa-lyon.fr> 於 2023年2月13日 週一 下午5:40寫道：
>>
>>> 1) This is for non-rigid motion. If the motion is rigid, it can be
>>> compensated by adjusting the source and detector positions and orientation
>>> wrt to the object in the geometry file.
>>> 2) Again, detector parallel to isocenter is meaning less, the isocenter
>>> is a point. And the goal of the geometry description is no to "make them
>>> parallel" but to account for the actual geometry of the system wrt scanned
>>> object.
>>>
>>> On Mon, Feb 13, 2023 at 9:44 AM 何明哲 <m10512067 at yuntech.org.tw> wrote:
>>>
>>>> Hi Simon
>>>> Thanks for reply，
>>>>
>>>> 1) I understand what you mean, I will be check out geometric parameter
>>>> again , but if there is a motion blur caused by the motion of the gantry,
>>>> is there any suggested way to correct it? I have reference to
>>>> RTK/Examples/MCCBCTReconstruction
>>>> <https://wiki.openrtk.org/index.php/RTK/Examples/MCCBCTReconstruction#Deformation_vector_field>,
>>>> but this is an example of the motion signal of the chest cavity, If it is
>>>> dental CBCT case, which file or format should the motion signal be in? Or
>>>> is there a document for reference?
>>>>
>>>> 2) Thanks for your reply, I understand, because the source is
>>>> considered as a point, so if the detector is not parallel to the ISO
>>>> Center, I just adjust the  out of plane angle to make them parallel, am I
>>>> right?
>>>>
>>>> The picture below is
>>>>
>>>> [image: 圖片1.bmp]
>>>>
>>>> thanks for your reply,
>>>> BR,
>>>>
>>>>
>>>>
>>>> Simon Rit <simon.rit at creatis.insa-lyon.fr> 於 2023年2月13日 週一 下午2:32寫道：
>>>>
>>>>> Hi,
>>>>> 1) I don't think this is due to detector blur. It seems to be a
>>>>> geometric problem. But it is hard to adjust it manually, this is generally
>>>>> calibrated.
>>>>> 2) I'm not sure I understand what your mean by "parallel to the
>>>>> source" as the source is assumed to be a point. If you only change SID,
>>>>> SDD, the 2 source and detector X/Y offsets, the Gantry Angle and the in
>>>>> plane angle, the detector is parallel to the rotation axis. The out of
>>>>> plane angle is the angle between the axis and the detector.
>>>>> Simon
>>>>>
>>>>> On Fri, Feb 10, 2023 at 12:02 PM 何明哲 <m10512067 at yuntech.org.tw> wrote:
>>>>>
>>>>>>
>>>>>> Hi All
>>>>>>
>>>>>> I have use the CudaFDKConeBeamReconstructionFilter  algorithm to
>>>>>> reconstruct a set of CBCT images,
>>>>>> 1. )but I found that each slices contains blurry image area , because
>>>>>> I was using a flat panel detector   , so I tried to adjust the source
>>>>>> offset , inofplaneangle and outofplaneangle , but the blurry image
>>>>>> area  still at there, could this be due to detector motion blur effect? Or
>>>>>> is there a suggested way to correct it?
>>>>>>
>>>>>> 2). From the reference http://www.openrtk.org/Doxygen/DocGeo3D.html I
>>>>>> can know that adjusting the source offset and inofplaneangle /
>>>>>> outofplaneangle can ensure that the detector remains parallel to the
>>>>>> source, But how can I make sure that the Z-axis of the ISO, the detector
>>>>>> and the source are also kept parallel at the same time, Can I achieve this
>>>>>> by adjusting any parameters or using any method?
>>>>>>
>>>>>> The picture below is my result  for Question 1.
>>>>>> [image: 圖片1.png]
>>>>>>
>>>>>> thanks for your reply,
>>>>>> BR,
>>>>>>
>>>>>> 回覆轉寄
>>>>>> _______________________________________________
>>>>>> Rtk-users mailing list
>>>>>> rtk-users at openrtk.org
>>>>>> https://www.creatis.insa-lyon.fr/mailman/listinfo/rtk-users
>>>>>>
>>>>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.creatis.insa-lyon.fr/pipermail/rtk-users/attachments/20230216/22945756/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ??1.png
Type: image/png
Size: 122517 bytes
Desc: not available
URL: <http://www.creatis.insa-lyon.fr/pipermail/rtk-users/attachments/20230216/22945756/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ??1.bmp
Type: image/bmp
Size: 324054 bytes
Desc: not available
URL: <http://www.creatis.insa-lyon.fr/pipermail/rtk-users/attachments/20230216/22945756/attachment-0001.bmp>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image200.png
Type: image/png
Size: 14762 bytes
Desc: not available
URL: <http://www.creatis.insa-lyon.fr/pipermail/rtk-users/attachments/20230216/22945756/attachment-0003.png>