M Lauria¹*, K Singhrao², J Lewis³, D O'Connell¹, W Lin⁴, A Santhanam¹, L Naumann¹, B Stiehl¹, P Boyle¹, P Lee⁵, D Low¹, (1) UCLA, Los Angeles, CA, (2) UCSF, San Francisco, CA, (3) Cedars-Sinai Medical Center, Pacific Palisades, CA, (4) Pepperdine University, Malibu, CA, (5) MD Anderson Cancer Center, Houston, TX

• 
  M Lauria

• 
  
• 
  
• 
  
• 
  
• 
  
• 
  
• 
  

PO-GePV-M-162 (Sunday, 7/10/2022)   [Eastern Time (GMT-4)]

ePoster Forums

Purpose: To investigate and document the variations in relative diaphragm position between the left and right lung across imaging sessions.

Methods: For 5 patients, we acquired 5DCT simulation models and cone-beam CT (CBCT) projections on the treatment day with a simultaneous respiratory bellows surrogate. To establish a calibration between amplitude signals across imaging sessions, we used the diaphragm position as a common reference. We deformed the 5DCT reference image to different amplitudes and used them to simulate projections. We established a calibration between the simulation and CBCT amplitudes by finding the amplitudes that aligned the diaphragms in each projection. We established these calibrations for the left and right lungs separately. Using the calibrated surrogates from the CBCT, we were able to obtain the diaphragm positions at end-exhalation and end-inhalation using the 5DCT model. To compare the relative diaphragm positions, we calculated the positions at end-exhalation for the left and right diaphragms in the simulation and CBCT scans. We recorded the difference between them and normalized the difference to the range of the diaphragm motion during CT simulation. We compared these differences across imaging sessions.

Results: For one patient case, the variation between left and right lungs was small as the normalized relative positions were -1.76 and -1.68. However, for the other 4 cases, these differences were more pronounced, with the most extreme difference being -0.017 during simulation and 0.791 during CBCT. For one patient, several fractions were acquired, and the relative diaphragm positions were 0.436, 0.456, and -0.602, indicating further variation across several days.

Conclusion: We have quantified variations in the left and right diaphragm positions from day to day. This has large implications on the reliability of respiratory motion modeling across several days and the overall stability of breathing patterns.

Funding Support, Disclosures, and Conflict of Interest: This work was funded by Varian (IRB Number 11-000620). Dr. Low has disclosed consulting work with ViewRay.

Keywords

Cone-beam CT, Motion Artifacts, Lung

Taxonomy

IM- Cone Beam CT: 4DCBCT

Contact Email