Sean Lauber:AMIRA - Generating Histograms

From OpenWetWare

AMIRA can be used to generate a histogram using a label field. In the previous example we generated a lung label by thoracic segmentation. Now we can generate a histogram of all the different densities and the proportion of these densities in this lung label. the data can then be exported to a .csv for use in Excel.

For CT scans the units of density are hounsfield units (HU). Here are some examples of different materials and their densities as HU:

-1000 HU: air
0: water
+1000: bone

1. Open the .raw file and the .AM file. In the example below the .am file is below the .raw file.

2. To link the label field back to the .raw file, click on the white square on the left of the .am file. Select ImageData and drag the line to the end of the .raw file, when you've connected the two files the line should turn light blue, now click the mouse to link the files.


4. Right click the linked .raw file, select Measure/Histogram.

5. Click the white square on the Histogram file and select Labels, then link the Histogram to the label file.


6. Click on the Histogram box and change the parameters as indicated to allow for 25 bins over the range -925 to 325 HU (range of 925+325 = 1250; 1250/25 bins = 50, so each bin represents the average of the counts over 50 HU). Be sure to select Material2 (the name of the Label Field)


7. Click Apply to generate the histogram.


Note: The X-axis is the density range (-1000 to +500 HU) and the Y-axis is the count. Each bin is

50 HU in size and there are 25 bins. You can see the first bin just after -625 HU and has a count

of about 0.05 x 10^5 (or 5000), which means that in this lung label there are 5000 voxels that have

a density of -625 HU. THere are two peaks, one on the left and the other on the right. The peak on

the left represents lung alveolar tissue (containing mostly air and therefore of low density) and

the peak on the right is the heart, diaphragm, and some connective tissue within the lung.

8. To export the histogram, File/Save Data, be sure to save as a CSV file to be accepted by Excel