A recent study conducted by a team of researchers from multiple sites around the world has shown that the use of deep-learning super-resolution algorithms in neuro MRI scans can substantially improve image quality and scan time.
iQMR a super-resolution AI-assisted algorithm was used to process low-resolution (LR) scans and produce enhanced-resolution (ER) images. These ER images were then compared to routine high-resolution (HR) scans, by a panel of experienced neuroradiologists.
The results of the study were positive, with the ER images exhibiting superior diagnostic quality, spatial resolution, noise levels, contrast resolution, and artifact appearance compared to the HR scans.
This study has important implications for the field of neuroimaging, as it demonstrates that the use of this deep-learning super-resolution algorithm can enable MRI scan time shortening, without adversely affecting image and diagnostic quality. This can greatly benefit patients, as shorter scan times can lead to increased comfort and convenience.
If you missed the presentation of this study, we invite you to watch the video to learn more about the methods and results of this study.
