Back

The use of hippocampal grading as a biomarker for early MCI

Morrison, C.; Dadar, M.; Shafiee, N.; Collins, D. L.

2022-02-21 neurology
10.1101/2022.02.15.22271017 medRxiv
Show abstract

INTRODUCTIONChanges in the hippocampus are associated with both increased age and cognitive decline due to mild cognitive impairment (MCI) and Alzheimers disease (AD). Most studies have examined the association between hippocampal changes and episodic memory, with many reporting a relationship between hippocampal measurements and global cognition. However, these studies often find associations only in the later stages of cognitive decline. The goal of this study was to examine if hippocampal grading is associated with global cognition in cognitively normal controls (NC), early MCI (eMCI), late (lMCI), and AD, and whether such associations differ across diagnostic cohorts. METHODSData from 1620 Alzheimers Disease Neuroimaging Initiative older adults were examined in this study (495 NC, 262 eMCI, 545 lMCI, and 318 AD). Participants were included if they completed baseline MRI scans and the Alzheimers disease Assessment Scale (ADAS-13) and Clinical Dementia Rating - Sum of Boxes (CDR-SB) cognitive tests. Linear regressions examined the influence of hippocampal grading on cognitive scores. RESULTSLower global cognition (i.e., increased ADAS-13 scores) was associated with hippocampal grading scores in all cohorts, including normal controls. Lower global cognition (i.e., increased CDR-SB scores) was associated with hippocampal grading scores in lMCI and AD, but not in eMCI or NC groups. DISCUSSIONThese findings suggest that hippocampal grading is associated with changes in global cognition in NC, eMCI, lMCI, and AD depending on the cognitive test. Thus, hippocampal grading may be a useful measure that is sensitive to progressive changes early in the disease course.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.