Software for automated identification and segmentation of perivascular spaces on MRI

Summary

Enlarged perivascular spaces (ePVS) in the brain are linked to several diseases, but detection at the resolution of clinical MRI relies on manual visual assessment, which are subjective, time-intensive and prone to error. The software automates ePVS detection in several commonly acquired MRI contrasts to increase efficiency and objectivity in quantifying whole brain ePVS burden in vivo.

Technology Overview

Increased areas of enlarged perivascular spaces (ePVS) are associated with many prevalent diseases, including Alzheimer’s disease, cerebral small vessel disease, cerebral amyloid angiopathy and multiple sclerosis.  Analysis of ePVS is typically done by manual visual assessment of multiple MRI scans which is subjective and laborious.

Oregon Health and Science University researchers have developed a fully-automated method of ePVS detection, with the following features:

• Objective identification of ePVS for improved accuracy and reproducibility.
• Faster ePVS analysis, as compared to manual assessment, allowing for rapid assessment of large datasets.
• Object-based morphometric estimates of each ePVS, providing more data to end-users. 
• Compatible with standard clinical field-strength (3 Tesla) MRI scans, highlighting a large potential market of both clinical and research-focused MRI users.

Publications

Schwartz et al., “Autoidentification of perivascular spaces in white matter using clinical filed strength T1 and FLAIR MR imaging” Neuroimage 202(2019). Link

Piantino et al., “Characterization of MR Imaging –Visible Perivascular Spaces in the White Matter of Healthy Adolescents at 3T.” Am J of Neuroradiology 41(2020): 2139. Link

Piantino et al., “Link between Mild Traumatic Brain Injury, Poor Sleep, and Magnetic Resonance Imaging: Visible Perivascular Spaces in Veterans.” J Neurotrauma 38(2021):2391. Link

Licensing Opportunity

This technology is available for licensing.