Protocols for accesssing information
Two files = one image
header.ascii
image.bin
An additional file named
image.bin.Z
will be included for convenience. The contents of image.bin.Z are redundant: It contains a compressed version of image.bin. It is produced with the UNIX “compress” command. The UNIX compression algorithm is lossless and can be reversed with the UNIX “uncompress” command. This file is provided so that file transfer time can be reduced for those with access to the UNIX compression programs.
The voxel file
one complete slice is displayed. The first voxel of the second slice follows the last voxel of the first slice, etc.
1 2 3 4 5
6 7 8 9 10
11 12 13 14 15
16 17 18 19 20
21 22 23 24 25
26 27 28 29 30
(Once this canonical display orientation is understood, other display orientations are reasonable as long as they represent proper rotations from this basic orientation.)
The Header file
The information groups
1. Identifying Information
2. Patient Information
3. Acquisition Information
4. Relationship Information
5. Image Presentation Information
Detailed group descriptions
Identifying Information
Length to end: //the number of bytes in the rest of the header file
Modality: {MR, CT, PET}
Manufacturer: {Siemens} //the manufacturer of the scanners in this project
Institution ID: {
Comments: //a message to serve as a reminder that the images may not used outside this project.
Patient Information
Other Patient ID: //a serial number used only for this project.
Comments: //a message stating that “Other patient ID” is unrelated to the actual patient ID.
Acquisition Information
Contrast/Bolus agent: //The value is “none” if there is no contrast agent.
Scanning sequence: {SE} //This line is present only for MR; only SE, meaning “Spin Echo”, is used in this project.
Radionuclide: //This line is present only for PET; it is the name of the isotope administered.
Slice thickness: //The units are millimeters.
KVP: //This line is present only for CT; it is the tube voltage.
Repetition time: //This line is present only for MR; it is TR in milliseconds.
Echo time: //This line is present only for MR; it is TE in milliseconds.
Number of averages: //This line is present only for MR; it is the number of acquisitions without change in any parameters (to reduce noise).
Gantry tilt: //This line is present only for CT and PET; machine's tilt.
Exposure time: //This line is present only for CT; units are milliseconds.
Exposure rate: //This line is present only for CT; units are milliamperes.
Comments: {T1,PD,T2} //This line is present only for MR; it labels the image as T1, PD, or T2 weighted (as determined by the TR and TE values).
Relationship Information
Study: {1, 2, 3, 4} //a set of series of one modality
Series: {1, 2, 3} //a set of logically related acquisitions
Acquisition: {1, 2} //a single continuous gathering of data over time
Patient Orientation: {L : P : H} //indicates three directions within the patient: toward the Left, toward the Posterior, and toward the top of the Head, respecti
Source image IDs: //This line is present only for rectified MR; it gives the Study\Series\Acquisition for each of two images, one with forward and one with reverse readout gradients. These are the two images that are used to construct this image.
Modified image description: {Rectified image} //This line is present only for rectified MR.
Comments: //This line is present only for rectified MR; it states that the image is a composition of two acquisitions.
Only the following nine combinations of Study\Series\Acquisition will be present in the database of images in this project:
3\2\1 = MR-PD
3\3\1 = MR-T2
4\2\1 = Rectified MR-PD
4\3\1 = Rectified MR-T2
----------------------------
| 1 2 3 4 5 |
Right | 6 7 8 9 10 | Left
| 11 12 13 14 15 |
----------------------------
Posterior
Display of slice 2, which is closer to the top of the head---
Anterior
---------------------------
| 16 17 18 19 20 |
Right | 21 22 23 24 25 | Left
| 26 27 28 29 30 |
--------------------------
Image Presentation Information
Rows := {128,256,512} //Number of rows in each slice
Columns := {128,256,512} //Number of columns in each slice
Slices := //Number of slices in the volume
Pixel size := 1.250000 : 1.250000 //The units are millimeters.
Compression code := //The .Z extension indicates compression via the Unix compress command, employs adaptive Lempel-Ziv coding.
Bits allocated := {16} //Before compression
Bits stored := {16} //Before compression
High bit := {15} //Most significant bit;
Pixel representation := {1} //1 means two's complement integer
Smallest pixel value := -65,536 //equals -2^16
Largest pixel value := +65,535 //equals +2^16 - 1
|--Study 1--(CT)
| |
| |
| |--Series 1--
| |
| |--Acquisition 1* [1\1\1]
|
|
|--Study 2--(PET)
| |
| |
| |--Series 1--
| |
| |--Acquisition 1* [2\1\1]
|
|
|--Study 3--(MR)
| |
| |
| |--Series 1--(T1)
| | |
| | |--Acquisition 1* (Forward gradient) [3\1\1]
| | |
| | |--Acquisition 2 (Reverse gradient) [3\1\2]
| |
| |
| |--Series 2--(PD)
| | |
| | |--Acquisition 1* (Forward gradient) [3\2\1]
| | |
| | |--Acquisition 2 (Reverse gradient) [3\2\2]
| |
| |
| |--Series 3--(T2)
| |
| |--Acquisition 1* (Forward gradient) [3\3\1]
| |
| |--Acquisition 2 (Reverse gradient) [3\3\2]
|
|
|
|--Study 4--(Rectified images derived as composites from Study 2)
| |
| |
| |--Series 1--(T1)
| | |
| | |--Acquisition 1* (from 3\1\1 and 3\1\2) [4\1\1]
| |
| |
| |--Series 2--(PD)
| | |
| | |--Acquisition 1* (from 3\2\1 and 3\2\2) [4\2\1]
| |
| |
| |--Series 3--(T2)
| |
| |--Acquisition 1* (from 3\3\1 and 3\3\2) [4\3\1]
|
|--Study 5--(MP-RAGE)
| |
| |
| |--Series 1--
| | |
| | |--Acquisition 1* [5\1\1]
FIGURE 1: The hierarchical relationship of Study, Series, and Acquisition number and their meanings.
Only the acquisitions with asterisks are included in the project.