Using a 3-T MRI scanner with a high gradient magnetic field (100mT/m), we evaluated the depiction of the intracranial fine branches on high-resolution 3D-TOF MRA (HR-TOF). We also assessed whether depiction can be improved by deep learning reconstruction (DLR).
Ten healthy volunteers were imaged by HR-TOF with DLR, and the sharpness of origin and the overall depiction of branches were assessed.
SNR, the sharpness of the origin and the overall depiction of branches were superior in HR-TOF with DLR.
HR-TOF can well depict fine branches from major trunks. By performing DLR processing, depiction can be improved.