Click here to


Are you sure ?

Yes, do it No, cancel

Reconstruction of a Lung Motion Model Using the Displacement Vector Fields Obtained From the Deformable Image Registration From CT Images

N Alsbou1*, S Ahmad2 , I Ali2 , (1) University of Central Oklahoma, Edmond, OK, (2) Oklahoma university Health Science Ctr., Oklahoma City, OK,


(Sunday, 7/29/2018) 4:00 PM - 4:30 PM

Room: Exhibit Hall | Forum 6

Purpose: To develop a model that reconstructs a voxel-by-voxel motion trajectory for mobile organs such as lungs using the displacement-vector-field (DVF) calculated by deformable image registration (DIR) algorithms using cone-beam-CT (CBCT) images.

Methods: The CBCT images for a thorax phantom were acquired for a mobile thorax phantom using kV-on-board imager mounted on a Varian Trilogy machine. The thorax phantom included three targets (small, medium and large) surrounded by low-density foam to simulate lung tissue. The phantom was moved sinusoidally with different motion amplitudes and frequencies. The CBCT images of the mobile phantom were registered to the stationary phantom images using different deformable image registration algorithms that include: (a) Demons, (b) Fast-Demons, (c) Horn-Schunck, and (d) Lucas-Kanade from the DIRART research software.

Results: The displacement vector fields obtained from the different DIR-algorithms from the respiratory motion were used to reconstruct a motion model for all voxels in the phantom. DVF represented the shifts of the individual voxels in the mobile tissue which were the positions of the different voxels at a certain time captured by the CBCT images. The maximal and minimal DVF correlated linearly with the motion amplitude of the mobile phantom for each target using the different DIR algorithms. The distribution of the DVF of voxels was used to extract the position-time relationship for each voxel and motion frequency. The current CBCT images were acquired over multiple respiratory cycles and thus the motion phase information was lost.

Conclusion: The study demonstrated the use of the DVF from DIR-algorithms to reconstruct the motion trajectory on a voxel-by-voxel basis for the lungs in a thorax phantom during respiratory motion. This lung motion model has potential applications in motion management of lung cancer patients treated with radiation therapy through the processes of treatment planning, patient setup and tumor localization and dose delivery.


Not Applicable / None Entered.


Not Applicable / None Entered.

Contact Email