Rohit Bakshi

This study shows that ODC1 suppresses age-associated B cell features in meningeal inflammation of MS and EAE, affecting disease severity.

The paper introduces the PARE framework to remove confounders from dimension reduction methods, showing it can eliminate batch effects in single-cell data and separate clinical from technical variability in neuroimaging.

This study shows that brain MRI volumetric biases persist across different scanners in a harmonized MS study, with site explaining over 50% of volume variation.

This paper develops a method to reduce technical variation in multisite MRI data by aligning intensity distributions across scanners.

TAPAS is an automated thresholding algorithm for segmenting brain lesions in MS using probability maps, reducing error and bias compared to manual and group-level thresholding.

The paper introduces a statistical technique to count distinct multiple sclerosis lesions from MRI data, showing it correlates with gold-standard counts and provides disease-related insights.