ESTRO 2024 - Abstract Book

S3904

Physics - Image acquisition and processing

ESTRO 2024

Figure 2: Weighted logarithmic subtraction for the liver target.

Figure 2 presents the result of the weighted logarithmic subtraction using a combination of different weighting factors and different registration types. The median CNR across all tested weighting factors was 2.0, 1.9, and 1.7 for deformable, rigid and no registration, respectively. The difference was statistically significant, with p-values equal to 7.7 x 10-9, 3.8x10-9 and 7.1x10-15 for deformable vs. rigid, deformable vs. no registration and rigid vs. no registration, respectively. This confirms the importance of correcting for motion in between frames, with improved results for the deformable approach [3].

Conclusion:

The integrated infrared tracking device provided accurate motion detection within 0.3-0.4 mm, to be used as benchmark in markerless tracking studies based on DE imaging. The weighting logarithmic subtraction factor equal to 0.8 along with deformable registration provided the best result in optimizing the CNR and suppressing the bony anatomy. Future work will include the implementation of noise reduction techniques to further enhance the image quality in DE images. Also, techniques for the automated selection of the optimal weighting factors will be pursued, and benchmarked against the motion detected via infrared tracking.

Keywords: DE imaging , markerless motion tracking , CBCT

References: 1. Patel R, Panfil J, Campana M, Block AM, Harkenrider MM, Surucu M, Roeske JC. Markerless motion tracking of lung tumors using dual-energy fluoroscopy. Med Phys. 2015 Jan;42(1):254-62. doi: 10.1118/1.4903892 2. Brody WR, Butt G, Hall A, Macovski A. A method for selective tissue and bone visualization using dual energy scanned projection radiography. Med Phys. 1981 May-Jun;8(3):353-7. doi: 10.1118/1.594957 3. Palaniappan P, Steininger P, Messner IM, Deutschmann H, Parodi K, Landry G, Riboldi M. "Motion Compensation