Automated sALPS: Advancing Non-Invasive Glymphatic Imaging for Early Cognitive Impairment Diagnosis and Staging

Motivation: Existing glymphatic biomarkers like ALPS are limited by ROI constraints and variability, requiring a more reliable, non-invasive tool for early diagnosis. Goal(s): This study aims to establish a more comprehensive biomarker for staging cognitive impairment by developing an automated, multi-ROI spectrum of ALPS (sALPS) analysis that mitigates observer bias and increases diagnostic sensitivity. Approach: We analyzed sALPS mean and variance from diffusion MRI data in a large cohort, correlating these metrics with plasma biomarkers and applying machine learning for classification. Results: Our sALPS indices demonstrated significant differentiation between cognitive stages, correlating strongly with plasma biomarkers and effectively classifying SCD and AD groups. Impact: This study provides a novel, automated sALPS biomarker that enhances early detection of glymphatic dysfunction, potentially improving diagnostic precision and staging of cognitive impairment through its application in machine learning models for robust classification across cognitive stages.