I'm collecting my DTI data on a Philips 3T with the following gradient table (see below).
To obtain proper vectors, the x-direction needs to be inverted. However, I'm not sure
how to handle the final direction in the table (33: 100,100,100). According to Philips, this
is "fake" data and can be ignored-thus I assumed it wouldn't play a role in the tensor
calculation. However, I've run BEDPOST with this value positive and negative (with all
other x-direction gradient values inverted) and I get different results.
I then ran BEDPOST with the gradients un-inverted and used fslmaths to invert the th and
ph merged_samples files and reran make_dyadic_vectors. The results were close to both
previous BEDPOST results with the inverted x-direction (with positive 100 and negative
100 values) but not an exact match for either one.
Do you have any advice for how to format this data for the bvecs file? Does it need to be
normalized as well?
Thanks in advance for your help!
-Laura
0: 1.000, 0.000, 0.000
1: 0.000, 1.000, 0.000
2: 0.000, 0.000, 1.000
3: -0.042, -0.115, -0.993
4: 0.1575, -0.201, -0.964
5: 0.232, -0.163, -0.959
6: 0.368, 0.026, -0.930
7: 0.190, 0.374, -0.908
8: -0.117, 0.833, -0.540
9: -0.200, 0.253, -0.947
10: -0.496, 0.134, -0.858
11: -0.014, -0.628, -0.778
12: -0.744, -0.148, -0.651
13: -0.761, 0.320, -0.564
14: -0.181, 0.925, -0.335
15: -0.680, -0.422, -0.600
16: 0.777, 0.471, -0.418
17: 0.924, -0.104, -0.368
18: 0.468, -0.767, -0.438
19: 0.882, -0.189, -0.432
20: 0.690, 0.706, -0.157
21: 0.239, 0.757, -0.608
22: -0.058, 0.984, 0.170
23: -0.537, 0.836, -0.113
24: -0.992, -0.121, -0.042
25: -0.997, 0.071, -0.038
26: -0.872, 0.478, -0.101
27: -0.249, 0.934, 0.258
28: 0.118, 0.992, -0.047
29: 0.338, 0.841, 0.422
30: 0.529, 0.841, 0.116
31: 0.997, 0.055, -0.057
32: 0, 0, 0
33: 100, 100, 100
|