[MINC-users] Differences between mincstats and volume_stats

[Andre MATOS]

I was comparing the results from mincstats and the volume_stats and I 
found some different values, specially in Median. The mincstats was 
run into a Mac and the volume_stats into Linux. Below are both:

mincstats sanca6_MTe2.mnc.gz -mask sanca6_2003_09_22_bayes1.mnc.gz 
-mask_range 4,4 -all
File:              sanca6_MTe2.mnc.gz
Mask file:         sanca6_2003_09_22_bayes1.mnc.gz
Total voxels:      3276800
# voxels:          158878
% of total:        4.848571777
Volume (mm3):      454553.6041
Min:               -0.02046138374
Max:               61.16808724
Sum:               5805114.165
Sum^2:             213541256.4
Mean:              36.53818757
Variance:          9.018965791
Stddev:            3.003159302
CoM_voxel(z,y,x):  28.03302643 142.6857334 128.9361516
CoM_real(x,y,z):   -1.402491803 40.19036803 6.234843781

Histogram:         (null)
Total voxels:      3276800
# voxels:          158878
% of total:        4.848571777
Median:            241.5912838
Majority:          37.16630936
BiModalT:          61.20598221
PctT [  0%]:       0
Entropy :          6.305040492

Bus error

-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-

volume_stats sanca6_MTe2.mnc.gz -mask sanca6_2003_09_22_bayes1.mnc.gz 
-maskrange 4 4 -all
File:         sanca6_MTe2.mnc.gz
Mask:         sanca6_2003_09_22_bayes1.mnc.gz
Mask extrema: [4, 4]
# voxels:     158878
% of total:   4.84857
Volume (mm3): 454554
Min:          -0.0271893
Max:          61.1572
Sum:          5.8051e+06
Sum2:         2.1354e+08
Mean:         36.5381
Variance:     9.01985
Stddev:       3.00331
Median:       36.8731
Majority:     37.1616
BiModalT:     35.8467
PctT (-1.79769e+308):    -0.0346599 61.1647
Entropy:      7.33663
CoM_voxel:    zspace:28.033 yspace:142.686 xspace:128.936
CoM_world:    zspace:6.23486 yspace:40.1904 xspace:-1.40249


Is anyone tryed to compile the volume_stats (N3 package) into a Mac OS X 
and also know which program (mincstats or volume_stats) is right?

--
Andre Matos  
amatos@mrs.mni.mcgill.ca



-------------
B
BTW, has anyone tested using mincstats as a drop-in replacement - can
one just symlink volume_stats to mincstats and scripts work properly
(producing the same output)?

            Peter
----
            Peter Neelin (neelin@bic.mni.mcgill.ca)