Automatic Quantification of Enlarged Perivascular Space in Patients with Traumatic Brain Injury Using Super-resolution of T2-weighted Images

The perivascular space (PVS) is important to brain waste clearance and brain metabolic homeostasis. Enlarged PVS (ePVS) becomes visible on MRI and is best appreciated on T2-weighted (T2w) images. Quantification of ePVS is however challenging because standard-of-care T1-weighted (T1w) and T2w images are often obtained as 2D acquisition, whereas accurate quantification of ePVS normally requires high-resolution volumetric 3D T1w and T2w images. The purpose of this study is to investigate the use of a deep learning based super-resolution technique to improve ePVS quantification from 2D T2w images for application in traumatic brain injury (TBI) patients. We prospectively recruited 26 volunteers (Age: 31±12yrs, 12M/14F) where both 2D T2w and 3D T2w were acquired along with 3D T1w images to validate the ePVS quantification using super-resolution T2w images. We then applied the super-resolution method to retrospectively acquired 2D T2w images in 41 chronic TBI patients (Age: 41±16yrs, 32M/9F). ePVS volumes were automatically quantified within the whole brain white matter and major brain lobes (temporal, parietal, frontal, occipital) in all subjects. Pittsburgh Sleep Quality Inventory (PSQI) was obtained on all TBI patients. Compared to the Silver Standard (3D T2w), in the validation study, the super-resolution T2w provided similar whole brain white matter ePVS volume (r=0.98, p