4254

View Abstract / View Presentation

Respiratory Motion Detection and Reconstruction Using CAPTURE and Deep Learning for a 0.35T MRI-LINAC System: An Initial Study

Sihao Chen¹, Cihat Eldeniz¹, Weijie Gan¹, Ulugbek Kamilov¹, Deshan Yang¹, Michael Gach¹, and Hongyu An¹
¹Washington University in St. Louis, Saint Louis, MO, United States

A self-navigated respiratory motion detection (CAPTURE) and a deep learning 4D reconstruction method were used to derive the 3D deformable motion field for a 0.35 T MRI-LINAC system. Promising results were obtained despite the low SNR at 0.35 T.

Figure 3. Images reconstructed using non-MoCo, MCNUFFT, CS and P2P from a healthy subject. The top row shows phase 1, corresponding to the end-of-expiration, and the bottom row shows phase 5, corresponding to the end-of-inspiration. Blue arrows point to a region of comparison for image quality among the three methods.

Figure 5. Images reconstructed using P2P at 0.35T and at 3T for different subjects for a rough qualitative assessment.