Question

Question Posted 05/04/15:
Dear experts,

I have recently noticed that there is a 4th pre-processing steps for T1 images (after gradwarping, B1 correction and N3 correction) called scaling. It's those images that are in MPRAGEMETA. I have not found information on this (except [here](http://adni.loni.usc.edu/scaled-2-uploads/) for the explanation of the scaled_2 images).

The first 3 steps (which may be optional depending on the scanner) are well described [here](http://adni.loni.usc.edu/methods/mri-analysis/mri-pre-processing/).

What about scaling? If I understand correctly, it uses a phantom. Is it the image labeled "B1 calibration PA"? It could be useful to update the information [here](http://adni.loni.usc.edu/methods/mri-analysis/mri-pre-processing/).

By the way, if I'm correct, the B1 correction uses a phantom too (I guess it's the image labeled "B1-Calibration Body").

Thanks in advance,
Mathieu
Response posted 08/06/15 by ADNI MRI Core:
Thank you for your question to the ADNI MRI Core.

The scans labeled “B1 Calibration PA” and “B1 Calibration body” were used to correct for the inhomogeneity of the image caused by the multichannel head coils.
The B1 Calibration PA- is the scan used to record the collection of image data via the Phased Array head coil connected to the scanner.
The B1 Calibration Body- records the image data with the built-in Body coil of the scanner. The scanning parameters for both the B1 scans are the same the only difference is the receive coil used.
The Phased Array scan will have good sensitivity near the actual individual coils of the array. As distance is increased from each coil the sensitivity drops dramatically and there is significant image inhomogeneity over the entire field of view.
The body coil has low sensitivity (compared to the best of the individual coils of a phased array) but the sensitivity of the coil is reasonably consistent over the entire field of view, this produces a fairly homogeneous image.
By acquiring a homogeneous body coil scan and comparing it to the inhomogeneous phased array scan we were able to calculate a correction for the inhomogeneity of the phased array coil scan. These B1 calibration scans were (supposed to be) completed on the subjects during their exam, for scanners that used multi-channel receive coils and did not have a vendor provided phased array inhomogeneity calibration scan(s).

The phantom scaling was achieved by scanning a completely separate exam from the subject. The site’s phantom was supposed to be scanned immediately after the subject scan and was labeled as a phantom. The resultant scans were labeled as “Scaled”.

B1 calibrations scans were acquired on both subject and phantom scans again, where the site used a multichannel phased array receive coil and did not have a vendor supplied inhomogeneity correction.

ADNI MRI