DTI metrics not computing correctly with b=0, 1000

[pajak]

Hello,

I have an issue with DTI.
I have data with 99 directions (b= 0, 1000, 2000) and I run it through the standard mrtrix pipline (denoise, degibbs, dwifslpreproc, biascorrect-ants) and then I extract only b0 and b1000 values in order to compute the tensor and metrics. None of the steps displays an error nor a warning. However, when I look at the metric output for MD, and RD it seems weird, the values are there but the image i just gray. On the other hand FA map looks fine and when I use the whole dataset (b0,b1000,b2000) the images look how they are supposed to.

I provide the dwgrad of the extracted image (b0, b1000)

(base) ➜ DWI mrinfo -dwgrad dwi_bval1000.mif
-0 0 0 0
-0 0 0 0
0.984191407393964 -0.0604681235296292 -0.166465851298979 1000.00094101227
0.559328072840852 0.64117585134683 0.525400451638374 1000.00031685393
0.35919633078325 0.886312303558669 -0.29228153638637 999.999752393266
0.469650469495491 -0.0429425353895877 -0.881807447891533 999.999780219203
-0.726861982710845 0.287769369604294 -0.623586760609305 1000.00082897016
0.360350909900797 -0.903188623201997 -0.233232786400516 1000.00011999558
-0.875719576599622 -0.226114152599902 0.426600062299816 1000.00020800086
-0.358728645807399 0.0862336336917787 -0.929450116519172 999.999680358746
0 0 0 0
-0.363526332985744 -0.92847445036359 -0.0760513001070176 999.99910867843
-0.370796559800497 0.507832713600681 0.777570476701042 999.99910729732
-0.138535659799656 -0.279006635599306 0.950243741497638 999.999480104971
0.81850641503935 -0.562394335827037 -0.117302427805639 999.99915050385
0.755553606899516 0.106016627199932 0.646451252499586 999.999598401282
-0.294646079706604 0.689619895915456 -0.661519528714826 999.999997955176
-0.259286226604826 -0.420667838207829 -0.86937289041618 1000.00015596278
0 0 0 0
-0.926446564448804 -0.360347726518983 -0.108840613805734 999.999468294643
-0.639449779588637 0.611628170289132 0.465848645691722 1000.00019003554
0.648081252619939 0.657949713820242 -0.383526875311799 1000.00078893847
0.563492230006705 -0.757346852409012 0.330003411903927 1000.0008659762
-0.0778956193708156 0.180018859101885 0.980574057810267 999.999427379059
-0.021277035300574 0.999510325026964 0.022943363600619 999.999059946045
0.806923483137003 -0.230675162210578 -0.543749447724934 1000.00098890829
-0.274620965094853 -0.664672160387542 0.694833969186977 1000.00024203749
0 0 0 0
-0.963895073078526 0.208534156295354 -0.165589231996311 1000.00033704456
0.658522350578743 0.66760340657845 0.347352854188787 999.999498764561
0.0941675177375462 -0.778253275079721 -0.620849674583822 999.999454952115
-0.718465262518868 -0.289833038507612 0.632300938116605 999.999695747476
0.0915094914511579 -0.844191601810682 0.528172843306683 1000.00015397469
0.807530563692668 -0.544449518695057 0.22686804599794 999.99969611816
-0.52482186321825 -0.454331252515799 -0.719822981625031 999.999631830452
-0 0 0 0
-0.453939307179834 0.324443226385587 -0.829864867463133 1000.00025808885
-0.747339209402427 -0.664286749602158 -0.0143951517100468 1000.0002499935
0.35671083629908 0.218177851799437 -0.908380869597656 1000.00065400516
0.52670721959209 -0.847107250487278 -0.0706315156389393 1000.00077203004
0.1339587074022 0.89748322271474 0.420212957506901 999.999068667153
-0.927301090619253 0.261208335305423 0.268109852305567 999.999201558475
0.155528880096962 0.916688597282092 -0.368093443392809 1000.00030203907
0.588163360026431 -0.220440597109906 -0.778119402834968 1000.00035791012
0 0 0 0
-0.918239395016033 -0.0730529157112755 -0.389229604406796 1000.00071296508
0.0618093593487484 -0.411685097391663 0.909227685281588 999.999809040501
-0.104655052197565 -0.607636499185863 -0.787289784581683 1000.00108204653
-0.892696949510333 -0.359728622704164 0.271454368803142 1000.00072297685
0.582349748525745 0.0560431266824776 0.811004277635854 999.999941311582
0.403853886215336 -0.798805433030333 -0.445883301716932 999.999289924053
0.268623229198245 -0.940844708093852 0.20652553349865 1000.00014701307
-0.196309669508609 0.383186258116804 0.902569003039581 999.999549312293
-0.466662905315683 -0.87955414952956 0.0927913295931185 999.999230232785
RD_map

MD_map

FA_map

FSL dtifit also does produce weird maps.

Scanner Siemens Prisma 3T

I apperciate all the help,

Paweł

[jdtournier]

Hi Pawel,

My guess is it’s simply a matter of setting the appropriate min/max intensity range for the display. I can’t see what it’s set to in your screenshots, they’re just off the top. With MRtrix, you should see values in the range 0 – 3×10^-3 s/mm² or so, but it’s not uncommon for one or two voxels to have really weird values (especially in the background noise) that cause the automatic determination of the display range to go wrong. I expect that’s all the problem is.

Cheers,
Donald.

[pajak]

That was the case.
Thank you!