next up previous contents
Next: KNN Classifier Up: Use of the DODECANTS Previous: Segmentation   Contents

CSF Volume Measurement

Once the registration and segmentation stages have been completed, the CSF volumes can be counted. Enter the directory and filename of the data list into the relevant fields. Then choose a filename for the output file that will contain the volume measurements, and enter it into the ``.tvs file" field. Ensure that the filename has the .tvs extension. Finally, press the ``Bulk CSF count" button. The tool will then run through the data sets listed in the data list, measuring the CSF volumes and some other important parameters, and output them to the .tvs file. This file will contain records that look like the following:

 alice 1 66.000000
 --a 9961.531250 --b 16343.578125  --c 9568.833008
 -+a 9009.450195 -+b 19038.785156  -+c 7531.889160
 +-a 9074.429688 +-b 18499.179688  +-c 9097.030273
 ++a 11967.398438 ++b 23259.582031  ++c 10958.814453
 x mm 122.108154, y mm 103.918434, z mm 157.280487

The first line consists of the label, disease code and age as given in the data list. The next four lines give the CSF volume measurements in each of the twelve sectors of the CSF space (in $mm^3$), in the order:



anterior left superior central left superior posterior left superior
anterior left inferior central left inferior posterior left inferior
anterior right superior central right superior posterior right superior
anterior right inferior central right inferior posterior right inferior

Left and right here refer to the image plane as seen on the monitor, not necessarily to the left and right hands of the subject. The final three lines give the maximal extents of the CSF space in the x (lateral), y (anterior-posterior) and z (inferior-superior) directions.

Many users will be interested only in overall CSF volume, and so the .tvs file can be rewritten into a format more suitable for reading into other analysis programs. Enter the directory and filename of the .tvs file into the ``directory" and ``.tvs file" fields of the DODECANTS tool and enter the desired output file name into the ``Output files" field. Then enter the number of records in the .tvs file (i.e. the number of data sets listed in the original data list) into the ``KNEAR" field, and press the ``TVS$->$V12 button. The .tvs file will be rewritten into columns in the output file, in the following order:

subject disease code
subject age
--a volume (normalised for head size)
--b volume (normalised for head size)
--c volume (normalised for head size)
-+a volume (normalised for head size)
-+b volume (normalised for head size)
-+c volume (normalised for head size)
+-a volume (normalised for head size)
+-b volume (normalised for head size)
+-c volume (normalised for head size)
++a volume (normalised for head size)
++b volume (normalised for head size)
++c volume (normalised for head size)
total CSF volume (not normalised)
CSF bounding box volume
total intracranial volume
total CSF volume (normalised)

The head size normalisation is accomplished through division by the total intracranial volume (TIV) [9]. The total intracranial volume is calculated from the volume of the bounding box on the CSF space, as described in TINA Memo no. 2004-002. Thus, all measurements are quoted as proportions of the TIV.


next up previous contents
Next: KNN Classifier Up: Use of the DODECANTS Previous: Segmentation   Contents
root 2017-09-26