next up previous contents
Next: Registration Up: Use of the DODECANTS Previous: Use of the DODECANTS   Contents

Preparation of the Data List

In order to allow rapid processing of large numbers of data sets, the DODECANTS tool requires a list of the data sets to be analysed as input. This list should be in text format and should look like the following:

No of data sets
Path to rough tair file
No of slices      Path to standard brain image
No of slices      Path to subject brain image
Label     Disease code      Age

``No of data sets" is the number of data sets in the list. ``Path to rough tair file" is the absolute path to the file containing the initial starting point for registration (see Section 3.2 for details). ``No of slices" and ``Path to standard brain image" are the absolute path to the MR image volume used to define the standard coordinate system and the number of slices in that volume (see Section 3.2 for details). Following these three lines, there are two lines for each subject in the data list: ``No of slices" and ``Path to subject brain image" are the number of slices in the subject MR image volume and the absolute path to that volume; ``Label" is a simple name that will be used to identify the subject in any output files; ``Disease code" is a single digit code (from 1 to 9) used to represent the disease that the patient is suffering from; and ``Age" is the age of the patient in years. An example is given below.

49 /home/pab/data/normal_brain
49 /home/pab/data/patient1
alice 4 61
49 /home/pab/data/patient2
bob 1 66

The label should be a text string with no spaces or special characters, and the disease code can be arbitrary: it will only be used to cluster the data in the KNN classifier and to prepare files listing the colours applied to each data point in the output files for display in xgobi. The colours applied are as followed:

1: SkyBlue
2: Orange
3: Red
4: Green
5: Yellow
6: Blue
7: White
8: White
9: Grey

Our in-house use of the labels is as follows:

1. Normal control
2. Alzheimer's disease
3. Vascular dementia
4. Fronto-temporal dementia
5. Lewy Body disease
6. Schizophrenia
7. Possible Alzheimer's disease
8. Possible Vascular dementia
9. Undefined

The DODECANTS tool requires that all of the data sets to be analysed are in ANALYZE format. However, TINA is capable of loading a variety of medical image formats and converting them. Therefore if, for example, you wish to analyse data in DICOM format, start a tinaTool, load each image volume into the Sequence tool, and output it as an ANALYZE image volume. Due to the restrictions of the ANALYZE format, the images must be cast to integer variable type prior to output: this can be accomplished by pressing the ``int" button on the ``image type" choice list in the sequence tool prior to output. See the TINA User's Guide for more information on the use of the Sequence tool

next up previous contents
Next: Registration Up: Use of the DODECANTS Previous: Use of the DODECANTS   Contents
root 2019-02-23