Brain Age: A novel approach to quantify the impact of multiple sclerosis on the brain (P1.371)
Citation Manager Formats
Make Comment
See Comments

Abstract
Objective: To evaluate a novel machine learning MRI model, and its ability to quantify ‘how much older’ a multiple sclerosis (MS) brain is than the patient’s chronological age.
Background: A predictive model of aging was previously built using machine learning in 1,537 healthy individuals, and can accurately predict chronological age based solely upon T1-weighted brain MRI (r=0.92). This brain age model can be applied to disease cohorts to quantify accelerated aging in diseased brains.
Design/Methods: Brain age was estimated using T1-weighted brain MRIs from 10 healthy controls (HCs) and 17 people with MS, both before and after 1 year of natalizumab treatment. Chronological age was subtracted from brain age, to give the ‘predicted age difference’ (BrainPAD) in each individual. This was also repeated after ‘lesion filling’ the MRIs.
Results: The MS brains were ‘10 years older’ than their chronological age (HC BrainPAD: −1.38±2.73; MS BrainPAD: +10.00±1.97; p=0.003). ‘Older brains’ correlated strongly with existing markers of disease severity in individuals (BrainPAD vs total lesion load; r=0.78, p<0.001). MS brains ‘got younger’ longitudinally after natalizumab in those who responded to treatment, versus those with ongoing disease activity (Change in BrainPAD: −0.67±0.41 vs +1.09±0.58; p<0.05). T1 lesions in MS contributed towards only 16% of the BrainPAD score.
Conclusions: MS brains were estimated to be far older than true chronological age. BrainPAD correlated with markers of disease severity and the ‘aging’ process was slowed by treatment. These findings are driven by MRI features other than focal MS lesions. The brain age model requires only a standard T1-weighted brain MRI, and can be applied to large cohorts with minimal user input. The machine learning taps into data beyond that possible with standard MRI analysis, and can produce results which are highly intuitive for researchers, clinicians and patients.
Disclosure: Dr. Raffel has received personal compensation for activities with GE Healthcare. Dr. Cole has nothing to disclose. Dr. Record has nothing to disclose. Dr. Sridharan has nothing to disclose. Dr. Sharp has nothing to disclose. Dr. Nicholas has received personal compensation for activities with Genzyme, Roche, Novartis, and Biogen.
Disputes & Debates: Rapid online correspondence
NOTE: All authors' disclosures must be entered and current in our database before comments can be posted. Enter and update disclosures at http://submit.neurology.org. Exception: replies to comments concerning an article you originally authored do not require updated disclosures.
- Stay timely. Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- 200 words maximum.
- 5 references maximum. Reference 1 must be the article on which you are commenting.
- 5 authors maximum. Exception: replies can include all original authors of the article.
- Submitted comments are subject to editing and editor review prior to posting.
You May Also be Interested in
Related Articles
- No related articles found.