General Information:

Screening/Baseline scans (Timepoint 1). The 1.5T MRI screening scan becomes the baseline scan for all subjects approved for the study. Therefore, when searching the database for baseline or "Timepoint 1" data, users should select both "screening" and "baseline" from the Visit selection box.

The ADNI exam consists of the following imaging sequences:
Subject Scan

1. Localizer/Scout Scan (20 secs)
2. Straight Sagittal 3D MPRAGE (8-10 mins)
3. Straight Sagittal 3D MPRAGE - REPEAT - (8-10 mins)
4. B1 Calibration Scan Phase Array Coil (if applicable) (30 secs)
5. B1 Calibration Scan Body Coil (if applicable) (30 secs)
6. Axial Dual Echo T2 FSE (5 mins)

ADNI Phantom - Quality Control Scans (following subject scan)

1. Localizer/Scout Scan (20 secs)
2. Straight Sagittal 3D MPRAGE (8-10 mins)

The MPRAGE is the T1-wieghted 3D series which will be used for most morphometric analyses. Two identical MPRAGE scans are done in each patient study back to back. The purpose of this is to maximize the probability that at least one high quality MPRAGE scan is obtained at each examination, and minimize the probability that subjects will need to return for a repeat study because of suboptimal image quality. The B1-calibration scans are obtained to correct for image intensity inhomogeneity due to receiver coil non uniformity, as described below, and would not be useful to the typical user. The dual fast spin echo (FSE) series a T2/proton density scan is nobtained for pathology detection.

Image Corrections Provided by ADNI
Each MPRAGE image in the database at LONI is linked with related image files which have undergone specific image preprocessing correction steps.  These corrections are as follows:

1. Gradwarp – gradwarp is a system specific correction of image geometry distortion due to gradient non-linearity.  The degree to which images are distorted due to gradient non-linearity varies with each specific gradient model. It is anticipated that most users would want to use images which have been corrected for gradient non-linearity distortion in analyses.
2. B1 non-uniformity – this correction procedure employs the B1 calibration scans noted in the protocol above to correct the image intensity non-uniformity that results when RF transmission is performed with a more uniform body coil while reception is performed with a less uniform head coil.
3. N3 – N3 is a histogram peak sharpening algorithm which is applied to all images. It is applied after grad warp and after B1 correction for systems on which these two correction steps are performed. N3 will reduce intensity non-uniformity due to the wave or the dielectric effect at 3T. 1.5T scans also undergo N3 processing to reduce residual intensity non-uniformity.

The need to perform the image preprocessing corrections outlined above does vary with manufacturer and system RF coil configuration. Philips Systems were equipped with B1 correction as product at the time ADNI began. In addition, Phillips gradient systems tend to be linear. Therefore, no gradwarped and no B1 corrected preprocessed files are generated for images acquired on Phillips Systems. The files available by manufacturer will be:

Phillips Systems:

1. unpreprocessed DICOM
2. N3 corrected

GE and Siemens systems with transmit-receive head RF coils:

1. unpreprocessed DICOM
2. gradwarped
3. gradwarp plus N3

GE and Siemens systems with receive-only head RF coils:

1. unpreprocessed DICOM
2. gradwarped
3. gradwarp plus B1 plus N3

As noted above it is anticipated that nearly every user would want to employ scans which have undergone gradwarp correction in analyses. Users that have developed their own set of tools for image intensity corrections may wish to simply use the gradwarped files. However, it is anticipated that most users will want to use the fully pre-processed files. These are most easily identified as files which contain N3 n the identifier. Note that these corrections are applied only to MPRAGE images (not FSE), and as outlined below only to the one MPRAGE volume associated with each time point that has been designated as “best” by the ADNI quality assurance team.

Phantom based scaling measures:
In addition to the corrections outlined above, phantom based measures of spatial scaling are associated with each MPRAGE image in an accompanying XML file. A version of the image with these spatial scale factors applied will be provided. Recall that each ADNI human exam is followed immediately by an acquisition with the ADNI phantom. Absolute scaling along each of the cardinal axes (x, y, z) is measured with the phantom. These phantom based measurements can be used to retrospectively scale the accompanying human MPRAGE image. In the limit that the image matrix is aligned with the cardinal axes, this amounts to adjusting the voxel size.  For images which this does not hold, application of the scale factors is slightly more complicated as scaling along one axis in the magnet will be mixed into the other two dimensions by the oblique rotation.

Use of image quality rating data in ADNI
Each exam undergoes a quality control evaluation at Mayo Clinic. Exams are evaluated for the presence of structural abnormalities that may affect inclusion/exclusion from the study at the time of baseline screening evaluation. Obviously, if a subject appears in the LONI data base, he/she has passed inclusion criteria. For example, the presence of white matter disease is not an exclusionary criteria for ADNI, but users may want to restrict download of scans to those which have little or no white matter disease. Users may want to reference the quality control results contained in the clinical download files when selecting scans for download based on presence or absence of certain abnormities on MRI.

The MRI quality control files also contains scan quality ratings. The presence, absence and severity of common artifacts (e.g. blurring due to head motion) is indicated. Contrast to noise and intensity homogeneity (after the ADNI corrections listed above) are also graded. Users may want to restrict download of scans to those which meet certain quality criteria. Using the criteria provided in the QC files, users can choose scans which met any desired set of image quality criteria for download from the MRI data base.

Finally, recall that back-to-back MPRAGE scans are acquired on each subject. The final step in scan grading is designation of one MPRAGE as “the” scan recommended for use. In the event of a repeat exam for a particular time point due to image quality problems, one of the four possible MPRAGE scans acquired is designated as “the” best MPRAGE for that particular time point. The ADNI preprocessing steps outlined above are applied only to the one MPRAGE designated as “best” at each time point. Obviously, the user can select whichever scan is desired, but it is anticipated that the scan selected as best at each time point by the ADNI MRI quality assurance team would be most useful to typical users.

Which Scan to select?
For the sake of completeness, the various files with different levels of pre-processing correction above are available to all users. However, it is envisioned that most users would want to use the scans that have undergone the maximum correction in their analyses.  This file will be the MPRAGE that has been identified as “best” in the quality ratings, and undergone gradwarping, intensity correction, and has been scaled for gradient drift using the phantom data. This will be identifiable as the file with “N3” and ”scaled” in the file name. 

Masks

Masks created by the MR Core as part of preprocessing are useful for performing N3 correction and are not brain masks. Therefore, the description of these masks has been updated to read "Intracranial Space" rather than "Brain."

PET

PET data from ADNI will be subject to a complex processing stream in order to bring images from multiple scanners into a common format, orientation, image uniformity, and spatial resolution.  Additional steps of this processing stream are planned.  The information provided below gives descriptions of the PET data that is currently available.

Raw PET image data:
Raw or “original” PET image data sets are available to approved investigators from this website.  These images have passed our quality control process.  However, we strongly advise against using these images in any analyses.  The images are from many different PET scanner models and differ in resolution, orientation, voxel and image dimensions, count statistics, etc. and cannot be aggregated without serious methodological limitations.  These images are in multiple different file formats, including DICOM, matrix file format (CTI-7) andInterfile format.

All PET scans are acquired using one of three different protocols:  1) dynamic: a 30 minute, six frame acquisition (6 five-minute frames), with scanning from 30 to 60 min post-FDG injection;  2) static: a single-frame 30 min acquisition with scanning 30-60 min post-injection (for Siemens PET/CT scanners that do not have dynamic scan acquisition capability);  and 3) quantitative: a 60 min dynamic protocol consisting of 33 frames, with scanning beginning at injection and continuing for 60 min that can be used to compute absolute glucose metabolic rate from a calculated from radioisotope input function measured in the carotid arteries.  The majority of the scans in the ADNI study were acquired with the first acquisition protocol.

Raw files can be found using either the “Simple Query” option of LONI’s image database search, or the “Advanced Query” selecting the “Original” option under “Image Information” section.

Processed PET image data:
Currently there are four types of processed PET image data available at LONI, each of which is described below.  All processed PET image data are in DICOM format.  These files can only be found by using the “Advanced Query” and selecting the “Processed” option under “Image Information” section of the search page.
The goals of these initial processing steps include 1) making more uniform PET data available, and 2) making PET images from different systems more similar.  Achieving these goals will provide consistent starting points and simplify subsequent ADNI analyses.
The processing steps listed below are incremental.  Scans from previous steps are further processed to provide increasingly uniform data sets.  Investigators should determine what type of processed data would be most appropriate for their planned use.  Scans acquired with dynamic and quantitative protocols have all four types of processed data.  Scans acquired with the static protocol have only two types of processed data (see below). 

1. Co-registered Dynamic:  Raw PET images from all sites are downloaded for quality control at University of Michigan.  Raw images are converted to a standard file format.  Separate frames are extracted from the image file for registration purposes.  In most cases, six (6) five-minute frames are acquired 30 to 60 minutes post-injection.  Each extracted frame is co-registered to the first extracted frame of the raw image file (frame acquired at 30-35 min post-injection).  The base frame image and the five co-registered frames (or all co-registered frames for the quantitative studies) are recombined into a co-registered dynamic image set.  These image sets have the same image size (for example, 128x128x63) and voxel dimensions (for example, 2.0x2.0x2.0 mm) and remain in the same spatial orientation as the original PET image data.  This is called ‘native’ space. These files are uploaded to LONI in DICOM format.   Only PET scans acquired under protocol 1 or 3 above, will have a processed image set of this type.   To summarize, PET image sets with a series description of “co-registered dynamic” have two primary differences from the “original” raw PET image data:  1) separate frames have been co-registered to one another lessening the effects of patient motion, and 2) image files are in DICOM format.
   
   
2. Co-registered, Averaged:  This type of processed image set is generated simply by averaging the 6 five-minute frames (or the last 6 frames for the quantitative studies) of the “Co-registered Dynamic” image set described above.  This creates a single 30 min PET image set still in “native” space.  As above, only PET scans acquired under protocol 1 or 3 above, will have an entry of this type.



3. Co-reg, Avg, Standardized Image and Voxel Size:  Each subject’s co-registered, averaged image from their baseline PET scan is then reoriented into a standard 160x160x96 voxel image grid, having 1.5 mm cubic voxels.  This image grid is oriented such that the anterior-posterior axis of the subject is parallel to the AC-PC line.  This is referred to as “AC-PC” space in the LONI search program.  This standardized image then serves as a reference image for all PET scans on that subject.  The individual frames from each PET scan (the baseline study as well as all subsequent studies (6-month scan, 12-month scan, etc.) are co-registered to this baseline reference image.  By doing the co-registration from the original raw image data to a standardized space in a single step, only one interpolation of the image data is required, and thus resolution degradation by interpolation is kept to a minimum, and is the same for all scans.  An averaged image is generated from the “AC-PC” co-registered frames and then intensity normalized using a subject-specific mask so that the average of voxels within the mask is exactly one.  Both the spatial orientation (AC-PC) and the intensity normalization of the image are intended as a starting point for subsequent analyses.  With a standardized image matrix, PET data from different scanner models can be compared more easily.  It should be noted that in these images sets only spatial re-orientation and intensity normalization of scans has occurred.  No non-linear warping or even linear scaling of the brain dimensions has been applied to the images.



4. Co-reg, Avg, Std Img and Vox Siz, Uniform Resolution:  These images are the result of smoothing of the above-mentioned images.  Each image set is filtered with a scanner-specific filter function (can be a non-isotropic filter) to produce images of a uniform isotropic resolution of 8 mm FWHM, the approximate resolution of the lowest resolution scanners used in ADNI.  Image sets from higher resolution scanners obviously have been smoothed more then image sets from lower resolution scanners.  The specific filter functions were determined from the Hoffman phantom PET scans that were acquired during the certification process.