Thank you for your answer
Now an other question:
If I have images with number stored in 16 bit short I should expect a grayscale from 0 to 65536 (2e+16); so why the grayscale range displayed is usually shorter (for example 0 8000) and different from an image to an other one? It could depend on the scanner image acquisition software, because afni by default copies exactly the input data files? And if it so, may be the image acquisition software arranges an array of consecutive graylevels in one graylevel, for example numbers from 0 to 8 are set 1, from 9 to 16 are set to 2 and so on, so at the end the gray scale derived is shorter than original one?
Thank you in advance for your answer