Removing Technical Variability in MRI Makes Discoveries More Replicable

Removing Technical Variability in MRI Makes Discoveries More Replicable

May 2016

The products of different scanners and imaging sites vary, meaning that structural MRI studies can have biases and can lack reproducibility. The investigators proposed a statistical method called RAVEL, which normalizes magnetic resonance imaging (MRI) intensities across scanners and imaging sites. RAVEL is a quick, easy method that allows imaging from multiple MRI scans to be quantitatively integrated, so that we can replicate new scientific discoveries.

The investigators modeled scanner effects, using a control region of the brain that is not associated with clinical covariates. They showed that RAVEL makes MRI scans more comparable in imaging databases. RAVEL promises increased sensitivity to disease-related changes; in particular, it makes the brain regions associated with Alzheimer’s disease in a large multi-center study more replicable.


Jean-Philippe Fortin, Elizabeth M. Sweeney, John Muschelli, Ciprian M. Crainiceanu,
Russell T. Shinohara, The Alzheimer's Disease Neuroimaging Initiative

About Us

To understand health and disease today, we need new thinking and novel science —the kind  we create when multiple disciplines work together from the ground up. That is why this department has put forward a bold vision in population-health science: a single academic home for biostatistics, epidemiology and informatics. 

© 2023 Trustees of the University of Pennsylvania. All rights reserved.. | Disclaimer

Follow Us