Search

Background: Geometric distortion is a serious problem in MRI, particularly in MRI guided therapy. A lack of affordable and adaptable tools in this area limits research progress and harmonized quality assurance. Purpose: To develop and test a suite of open‐source hardware and software tools for the measurement, characterization, reporting, and correction of geometric distortion in MRI. Methods: An open‐source python library was developed, comprising modules for parametric phantom design, data processing, spherical harmonics, distortion correction, and interactive reporting. The code was used to design and manufacture a distortion phantom consisting of 618 oil filled markers covering a sphere of radius 150 mm. This phantom was imaged on a CT scanner and a novel split‐bore 1.0 T MRI magnet. The CT images provide distortion‐free dataset. These data were used to test all modules of the open‐source software. Results: All markers were successfully extracted from all images. The distorted MRI markers were mapped to undistorted CT data using an iterative search approach. Spherical harmonics reconstructed the fitted gradient data to 1.0 ± 0.6% of the input data. High resolution data were reconstructed via spherical harmonics and used to generate an interactive report. Finally, distortion correction on an independent data set reduced distortion inside the DSV from 5.5 ± 3.1 to 1.6 ± 0.8 mm. Conclusion: Open‐source hardware and software for the measurement, characterization, reporting, and correction of geometric distortion in MRI have been developed. The utility of these tools has been demonstrated via their application on a novel 1.0 T split bore magnet. [ABSTRACT FROM AUTHOR]

Background: MRI-guidance techniques that dynamically adapt radiation beams to follow tumor motion in real time will lead to more accurate cancer treatments and reduced collateral healthy tissue damage.The gold-standard for reconstruction of undersampled MR data is compressed sensing (CS) which is computationally slow and limits the rate that images can be available for real-time adaptation. Purpose: Once trained, neural networks can be used to accurately reconstruct raw MRI data with minimal latency.Here,we test the suitability of deep-learningbased image reconstruction for real-time tracking applications on MRI-Linacs. Methods: We use automated transform by manifold approximation (AUTOMAP), a generalized framework that maps raw MR signal to the target image domain, to rapidly reconstruct images from undersampled radial k-space data. The AUTOMAP neural network was trained to reconstruct images from a golden-angle radial acquisition, a benchmark for motion-sensitive imaging, on lung cancer patient data and generic images from ImageNet. Model training was subsequently augmented with motion-encoded k-space data derived from videos in the YouTube-8M dataset to encourage motion robust reconstruction. Results: AUTOMAP models fine-tuned on retrospectively acquired lung cancer patient data reconstructed radial k-space with equivalent accuracy to CS but with much shorter processing times. Validation of motion-trained models with a virtual dynamic lung tumor phantom showed that the generalized motion properties learned from YouTube lead to improved target tracking accuracy. Conclusion: AUTOMAP can achieve real-time, accurate reconstruction of radial data. These findings imply that neural-network-based reconstruction is potentially superior to alternative approaches for real-time image guidance applications. [ABSTRACT FROM AUTHOR]