Hello,
I am wondering if using 3dclustsim or ETAC for cluster-sized thresholding is appropriate for determining significant voxels of diffusion MRI data (i.e., conventional metrics such as FA or MD, or more advanced metrics such as NODDI parameters). I have tried implementing the -Clustsim option into 3dttest++ and find very elevated cluster size threshold values (e.g., NN3-2sided cluster size of 900+ for p=0.05/p=0.005). See below table.
Is there a more appropriate method to use? Is using clustsim/ETAC for diffusion data discouraged?
Thanks in advance,
Jesse
# CLUSTER SIZE THRESHOLD(pthr,alpha) in Voxels
# -NN 3 | alpha = Prob(Cluster >= given size)
# pthr | .10000 .09000 .08000 .07000 .06000 .05000 .04000 .03000 .02000 .01000
# ------ | ------ ------ ------ ------ ------ ------ ------ ------ ------ ------
0.100000 12924 13721 14714 15705 16782 17848 19227 20862 22791 25912
0.090000 11333 12226 13014 13965 15068 16261 17527 19037 21116 24431
0.080000 9655 10352 11326 12152 13267 14356 15704 17151 19275 22704
0.070000 8067 8735 9407 10248 11232 12444 13713 15189 17149 20579
0.060000 6550 7182 7775 8486 9349 10390 11641 13276 15076 18648
0.050000 5108 5598 6061 6725 7543 8388 9382 11038 12998 16192
0.040000 3848 4201 4589 5056 5664 6341 7532 8747 10521 13630
0.030000 2745 2982 3285 3669 4115 4648 5413 6370 7810 10314
0.020000 1821 1963 2131 2333 2682 2994 3377 4067 4865 7019
0.015000 1400 1516 1634 1759 1947 2246 2603 3040 3826 5250
0.010000 977 1057 1139 1242 1379 1556 1797 2112 2662 3588
0.007000 727 787 859 938 1033 1165 1324 1569 1942 2545
0.005000 562 607 657 719 804 903 1022 1181 1486 2014
0.003000 383 409 451 499 554 635 712 825 1025 1378
0.002000 287 311 344 378 423 478 547 643 805 1067
0.001500 235 256 285 316 351 398 453 537 665 899
0.001000 176 195 217 241 267 302 354 407 518 689
0.000700 139 152 168 189 211 239 279 331 416 568
0.000500 109 121 134 150 170 194 223 268 338 465
0.000300 75 84 93 104 119 137 159 198 245 352
0.000200 55 62 68 78 89 101 121 150 194 281
0.000150 44 50 56 63 73 84 100 125 163 239
0.000100 31 35 41 47 54 63 75 95 126 191
7.000e-5 22 25 30 35 40 48 59 74 101 154
5.000e-5 16 19 22 26 31 37 46 58 82 126
3.000e-5 9 11 13 15 19 24 30 38 57 91
2.000e-5 5 6 8 10 12 16 22 29 42 69
1.500e-5 3 4 5 6 9 12 16 22 33 56
1.000e-5 1 2 2 3 5 7 10 15 24 43
