Abstract: Described is an approach for tracking 3D organ motion in real-time using magnetic resonance imaging (MRI). The approach may include offline learning, which may acquire signature and 3D imaging data over multiple respiratory cycles to create a database of high-resolution 3D motion states. The approach may further include online matching, which may acquire signature data only in real-time (latency less than 0.2 seconds). From a motion state and motion signature database, the 3D motion state whose signature best (or sufficiently) matches the newly-acquired signature data may be selected. Real-time 3D motion tracking may be accomplished by performing time-consuming acquisition and reconstruction work in an offline learning phase, leaving just signature acquisition and correlation analysis in an online matching step, minimizing or otherwise reducing latency. The approach may be used to adapt radiotherapy procedures based on tumor motion using a magnetic resonance linear accelerator (MR-Linac) system.

Abstract: Described is an approach for tracking 3D organ motion in real-time using magnetic resonance imaging (MRI). The approach may include offline learning, which may acquire signature and 3D imaging data over multiple respiratory cycles to create a database of high-resolution 3D motion states. The approach may further include online matching, which may acquire signature data only in real-time (latency less than 0.2 seconds). From a motion state and motion signature database, the 3D motion state whose signature best (or sufficiently) matches the newly-acquired signature data may be selected. Real-time 3D motion tracking may be accomplished by performing time-consuming acquisition and reconstruction work in an offline learning phase, leaving just signature acquisition and correlation analysis in an online matching step, minimizing or otherwise reducing latency. The approach may be used to adapt radiotherapy procedures based on tumor motion using a magnetic resonance linear accelerator (MR-Linac) system.