Brain age estimation is a sensitive marker of processing speed in the early course of multiple sclerosis
Einar A. Høgestøl,
Gro O. Nygaard,
Petter E. Emhjellen,
Mona K. Beyer,
Ole A Andreassen,
Elisabeth Gulowsen Celius,
Nils Inge Landrø,
Lars T. Westlye,
Hanne Flinstad Harbo
Posted 29 May 2019
bioRxiv DOI: 10.1101/651521
Posted 29 May 2019
Background and objectives: Cognitive deficits in MS are common, also early in the disease course. We aimed to identify if estimated brain age from MRI could serve as an imaging marker for early cognitive symptoms in a longitudinal MS study. Methods: A group of 76 MS patients (mean age 34 years, 71% females, 96% relapsing-remitting) was examined 1, 2 and 5 years after diagnosis. A machine-learning model using Freesurfer-processed T1-weighted brain MRI data from 3208 healthy controls, was applied to develop a prediction model for brain age. The difference between estimated and chronological brain age was calculated (brain age gap). Tests of memory, attention and executive functions were performed. Associations between brain age gap and cognitive performance were assessed using linear mixed effects (LME) models and corrected for multiple testing. Results: LME models revealed a significant association between the Color Naming condition of Color-Word Interference Test and brain age gap (t=2.84, p=0.005). Conclusions: In this study, decreased information processing speed correlated with increased brain age gap. Our findings suggest that brain age estimation using MRI provides a useful semi-automated approach applying machine learning for individual level brain phenotyping and correlates with information processing speed in the early course of MS.
- Downloaded 224 times
- Download rankings, all-time:
- Site-wide: 73,312 out of 100,715
- In neuroscience: 12,952 out of 17,942
- Year to date:
- Site-wide: 84,748 out of 100,715
- Since beginning of last month:
- Site-wide: None out of 100,715
Downloads over time
Distribution of downloads per paper, site-wide
- 20 Oct 2020: Support for sorting preprints using Twitter activity has been removed, at least temporarily, until a new source of social media activity data becomes available.
- 18 Dec 2019: We're pleased to announce PanLingua, a new tool that enables you to search for machine-translated bioRxiv preprints using more than 100 different languages.
- 21 May 2019: PLOS Biology has published a community page about Rxivist.org and its design.
- 10 May 2019: The paper analyzing the Rxivist dataset has been published at eLife.
- 1 Mar 2019: We now have summary statistics about bioRxiv downloads and submissions.
- 8 Feb 2019: Data from Altmetric is now available on the Rxivist details page for every preprint. Look for the "donut" under the download metrics.
- 30 Jan 2019: preLights has featured the Rxivist preprint and written about our findings.
- 22 Jan 2019: Nature just published an article about Rxivist and our data.
- 13 Jan 2019: The Rxivist preprint is live!