Hiring in life sciences? Share your open positions with our professional community. Read more Close

Advertisement

Harmonization of Longitudinal Multi-Scanner Neuroimaging Data With Nonlinear Biological Associations.

Created on 14 Nov 2026

Authors

Neel M Desai, Haochang Shou, Noah Hillman, Randa Melhem, Lori L Beason-Held, Susan M Resnick, Alzheimer's Disease Neuroimaging Initiative, Baltimore Longitudinal Study of Aging, Christos Davatzikos, Russell T Shinohara

Published in

Human brain mapping. Volume 47. Issue 16.

Abstract

The aggregation of neuroimaging data across multiple sources, typically scanners or imaging sites, has the potential to improve statistical power when conducting inference. The added benefits of increased sample size to inference, however, can be obstructed by unwanted source-specific technical variability that increases difficulty in estimating biological signals of interest. To address this issue, recent methods based on ComBat, a batch-correction procedure originally developed for genomics, have been developed for both cross-sectional and longitudinal neuroimaging data. While these approaches have successfully removed technical batch-specific variability, they restrict their consideration to linear biological associations. In this article we introduce a ComBat-based method to harmonize longitudinal multisource imaging data that can account for nonlinear biological associations of interest. We demonstrate via simulation that, in the presence of nonlinear biological signal, our method outperforms existing ComBat-based approaches designed for longitudinal imaging data, particularly in settings where observed values of relevant biological covariates lack substantial overlap across sources. Using volumetric measurements for 504 subjects combined from both the Baltimore Longitudinal Study of Aging (BLSA) and the Alzheimer's Disease Neuroimaging Initiative (ADNI), we identify the presence of both additive and multiplicative study-specific effects for a substantial number of brain regions and demonstrate the effective removal of their presence on data harmonized by our method. The flexibility of the proposed method towards capturing many types of biological associations makes it a versatile procedure that can be used to harmonize longitudinal biomedical data.

PMID:
43031254
Bibliographic data and abstract were imported from PubMed on 14 Nov 2026.

Read full publication at:
Please sign in to see all details.

Advertisement

Stats

  • Community rating n/a 0 votes
  • Reviewers' rating n/a 0 votes
  • Your rating

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

  • Recommendations n/a n/a positive of 0 vote(s)
  • Views 10
  • Comments 0

Recommended by

  • No recommendations yet.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.

Advertisement