1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
| %
% This script is intended to perform the full process
% of creating a rat-template based on the Paxonos atlas.
% Hence, it will also serve as a documentation of exactly
% what has been done.
%
% It will
% 1. Read a list of corresponding points defined by Petra
% in the Paxonos atlas and in a T2-weighted image volume
% of the rat we have chosen as our model rat. It will
% then find the affine transform that minimises the sum
% of squared distances between the paxonos and the T2
% point lists. Optionally it might minimise a weighted
% sum of squares where the weighting is derived from the
% subjective rating Petra supplied for each point.
%
% 2. Reslice the T2-weighted volume into a volume determined
% by Petra based on the extent of the Paxonos Atlas. It
% was determined we use a bounding box of -7mm->7mm in the
% x-direction, -2mm->10mm in the y-direction and 6mm->-16mm
% in the z-direction.
% For practical reasons I later extended the bounding box
% in the x and y directions, and decreased it in the z-direction
% such that the resulting bounding box was -8mm->8mm,
% -12mm->1mm (Bregma system) and -15.6mm->6mm (Bregma system).
% We also decided to go for 0.2x0.2x0.2mm voxels.
%
% This gives us reasonable matrix-sizes of 80x63x108 voxels.
%
% Furthermore it was decided (for practical reasons) to swap
% the y- and z-axes such that the "new" y-axis goes tail->nose
% in the rat, and the new z-axis goes belly->back in the rat.
% This way the aspect of the different projections is much more
% like those in the human brain, facilitating the adaptation to
% SPM.
%
% Also, it was decided that we work on tenths of mm, rather than
% in mm. This enables us to retain the one-to-one relationship
% between voxels in MIP.mat and the co-ordinates of the atlas,
% thereby preventing any major rewrite of spm_mip_ui.m.
%
% The result of this is that co-ordinate reported by SPM as
% [37 -45 -56] is to be interpreted as the position
% [3.7 -5.6 -4.5] (VLGPC, whatever that is) in the Paxinos
% Atlas.
%
% Two image volumes will be created, one from a filtered
% version (.8x.8x.8mm) of the T2-volume, to use as a template, and
% one from the unfiltered version to use as a "canonical" rat.
% There is also a brainmask.img file used to mask the
% spatial normalisation that is based on the canonical brain
% after applying a series of image-processing primitives
% (connected component labelling, dilatation and erosion).
% The files are
% /usr/local/spm99/ratlas/templates/canonical_T2.img
% /usr/local/spm99/ratlas/templates/template_T2.img
% /usr/local/spm99/ratlas/templates/brainmask.img
%
% 3. Create a very crude interface that allows a user to
% create a list of co-ordinates corresponding to outer
% contour in one selected slice each in axial, saggital
% and coronal sections of the Paxinaos atlas. This is
% based in .jpeg files supplied by Petra.
% These conturs have been saved to
% /d1/data/Petra/Paxinos.mat
% The contours were plotted to a 364x400 figure window,
% converted to a .bmp format and subsequently to a .mat
% format to render it compatible with MIP.mat.
% A variable named mip96 has been saved to
% /usr/local/spm99/ratlas/MIP.mat
%
% 4. Make necessary changes to spm_* routines. In order to
% facilitate adapatations to future releases of SPM etc
% it was decided to make as few and as small changes as
% possible to existing SPM routines. It was however not
% possible to avoid altogether.
% In general, each change has been "labelled" with a
% comment containing the word "Paxinos", so it should
% be easy to locate all altered lines.
% The following is a list of routines that has been
% changed, and a short description of the changes.
%
% /usr/local/spm99/ratlas/spm_project.c
% i) New values for widths and center-points of
% the mips.
% ii) The determination of wether a cluster falls
% within the borders of a mip is now based on
% atlas-coordinates rather than mip coordinates.
% iii) New way of determining coordinates of x-direction
% in transversal view.
%
% /usr/local/spm99/ratlas/spm_mip_ui.m
% i) New values for widths and center-points of
% the mips.
%
% /usr/local/spm99/ratlas/spm_sn3d.m
% i) It uses the Rat brainmask as default.
% ii) It uses a filter kernel of .8x.8x.8mm
% iii) It bases the calculation of FOV on transformed
% rather than original object coordinates.
%
% /usr/local/spm99/ratlas/spm_affsub3.m
% i) Disregards Bayesian prior based on human brains.
%
% /usr/local/spm99/ratlas/spm_defaults.m
% i) New value for sptl_Ornt that reflects zoom
% by a factor of 10, shift of y- and z-axes
% and the "non-central" nature of the Bregma.
% ii) New value for the bounding-box, reflecting
% the Paxinos atlas.
% iii) Spatial basis functions reflecting the aspect
% of the Rat brain.
%__________________________________________________________
% Jesper Andersson 23/1-02.
%
% As a part of facilitating the use of several rats
% conv_2_paxonos has been divided into several routines.
%
%
% When going through the data 14/6-02 the following
% observations were made.
%
% rat_plain_3:
% Worked fine. No outliers in the point distributions
% and a cursory check of normalised images vs the
% Paxinos atlas looked fine.
%
% rat_plain_4:
% Worked fine. No outliers in the point distributions
% and a cursory check of normalised images vs the
% Paxinos atlas looked fine.
%
% rat_plain_5 (aka spm_template):
% Worked fine. No outliers in the point distributions
% and a cursory check of normalised images vs the
% Paxinos atlas looked fine.
% %
% rat_plain_6:
% Both 'cpu' points were extreme outliers and disrupted
% the entire normalisation procedure. After removal of
% these it looked better, although the 'aci' points were
% now (not extreme) otliers. However, since these are unique
% in that they are the only points defining the mapping in
% the frontal bits I chose to leave them in.
%
% LFF4:
% Worked fine. No outliers in the point distributions
% and a cursory check of normalised images vs the
% Paxinos atlas looked fine.
%
%__________________________________________________________
% Jesper Andersson 14/6-02
%
% Here starts code pertaining to step 1 above.
%
%
% Start by reading .txt file.
%
%lecture des fichiers
txt_fname = '/d1/data/Petra/rat5/spm_template.txt';
ima_fname = '/d1/data/Petra/LFF4/s_0001.img';
hdr_fname = '/d1/data/Petra/LFF4/s_0001.hdr';
mat_fname = '/d1/data/Petra/LFF4/s_0001.mat';
crd = read_coordinates(txt_fname); %on recup les coordonnées ds un vecteur crd
%
% Remove cpu points for rat_plain_6.
%
%crd.spmc = [crd.spmc(1:23,:); crd.spmc(26:end,:)];
%crd.paxc = [crd.paxc(1:23,:); crd.paxc(26:end,:)];
%crd.qlty = crd.qlty([1:23 26:end]);
%j=1;
%for i=[1:23 26:length(crd.name)]
% crd.name{j} = crd.name{i};
% j = j+1;
%end
%
%
% Next create M matrix mapping between native and
% Paxinos space.
%
M = get_M(crd.spmc,crd.paxc,crd.qlty,crd.name);
%
% Save it as appurtenant .mat file.
%
save(mat_fname,'M');
%
% Create template (smooth) and canonical
% (non-smoothed) image volumes in
% Paxinos space.
%
make_tmplt_and_cnncl(ima_fname,mat_fname);
%
% Create (soft) average images for template
% and canonical brain. One of the images is
% quite different w.r.t. contrast compared
% to the others. Therefore I have choosen
% to perform a histogram-matching of all
% to the first (arbitratily chosen as
% rat_plain5).
%
%image canonique
cP = ['/d1/data/Petra/rat5/canonical_T2.img';...
'/d1/data/Petra/LFF4/canonical_T2.img';...
'/d1/data/Petra/rat3/canonical_T2.img';...
'/d1/data/Petra/rat4/canonical_T2.img';...
'/d1/data/Petra/rat6/canonical_T2.img'];
cP = spm_vol(cP);%on recup un vecteur de structures
cdata = spm_read_vols(cP);% cdata de la forme [ Y, XYZ], on aura en sortie(matrice 4D des données d'image)???
for i=1:length(cP)% pour chaque structure de cP
maxval(i) = max(max(max(cdata(:,:,:,i)))); %on prends le max sur cdata
cdata(:,:,:,i) = cdata(:,:,:,i) / maxval(i);% on normalise Cdata, dc la plus grande valeur pour chaque I sera 1
end
tmp = [];
%
for i=1:size(cdata,3)
tmp = [tmp cdata(:,:,i,1)];
end
nhist = hist(tmp(:),512); %retourne un vecteur de 10 cases à égal distance contenant les 512 éléments
hedata = zeros(size(cdata)); %matrice de zéros de la taille de Cdata
hedata(:,:,:,1) = cdata(:,:,:,1);
for i=2:5
tmp = [];
for j=1:size(cdata,3)
tmp = [tmp cdata(:,:,j,i)];
end
tmp = histeq(tmp,nhist);%contient les intensités modifiées ac les nhist degrés
for j=1:size(cdata,3)
hedata(:,:,j,i) = tmp(:,(j-1)*size(cdata,2)+1:j*size(cdata,2));%????
end
end
%
% And now create soft average
%
ssum = sum(hedata,4);%la somme des intensités
nn = sum(cdata>0,4); %la somme des voxels
s_ave = maxval(1)*ssum./nn; % la moyenne ??? pk multiplier par maxval(1)???
%
% Write this to template directory.
%
%on genère l'image canonique
oP.fname = '/d1/data/Petra/template/canonical_T2.img';
oP.dim = cP(1).dim;
oP.mat = cP(1).mat;
oP.pinfo = cP(1).pinfo;
oP.descrip = 'Canonical (5 rats) T2 Rat Brain in Paxonos Space';
oP = spm_write_vol(oP,s_ave);
%
% Do the same for the template
%
%template
tP = ['/d1/data/Petra/rat5/template_T2.img';...
'/d1/data/Petra/LFF4/template_T2.img';...
'/d1/data/Petra/rat3/template_T2.img';...
'/d1/data/Petra/rat4/template_T2.img';...
'/d1/data/Petra/rat6/template_T2.img'];
tP = spm_vol(tP);
tdata = spm_read_vols(tP);
for i=1:length(tP)
maxval(i) = max(max(max(tdata(:,:,:,i))));
tdata(:,:,:,i) = tdata(:,:,:,i) / maxval(i);
end
tmp = [];
for i=1:size(tdata,3)
tmp = [tmp tdata(:,:,i,1)];
end
nhist = hist(tmp(:),512);
hedata = zeros(size(tdata));
hedata(:,:,:,1) = tdata(:,:,:,1);
for i=2:5
tmp = [];
for j=1:size(tdata,3)
tmp = [tmp tdata(:,:,j,i)];
end
tmp = histeq(tmp,nhist);
for j=1:size(tdata,3)
hedata(:,:,j,i) = tmp(:,(j-1)*size(tdata,2)+1:j*size(tdata,2));
end
end
%
% And now create soft average
%
ssum = sum(hedata,4);
nn = sum(tdata>0,4);
s_ave = maxval(1)*ssum./nn;
%
% Write this to template directory.
%
oP.fname = '/d1/data/Petra/template/template_T2.img';
oP.dim = cP(1).dim;
oP.mat = cP(1).mat;
oP.pinfo = cP(1).pinfo;
oP.descrip = 'Template (5 rats) T2 Rat Brain in Paxonos Space';
oP = spm_write_vol(oP,s_ave);
%
% Now, lets try and create a decent brainmask.img
% for the template.
%
path('/home/jesper/matlab/PhaseMaps',path);
P = spm_vol('/d1/data/Petra/template/template_T2.img'); %vecteur de structures contenant les infos sur l'image
f = spm_read_vols(P); %
%
% Simple thresholding
%
msk = double(f>10.0e8);%seuillage
%
% Embed it in larger volume to allow
% for dilate and erode operations.
%
tmp = zeros(100,83,108);
tmp(11:90,11:73,:) = msk;
figure('Position',[050 050 800 800])
for i=1:100
subplot(10,10,i);
imagesc(tmp(:,:,i)>0.1);
axis('image'); axis off;
end
%
% Find largest connected component.
%%on cherche la plus grande composante connexe
[cctmp,crap] = flood_fill(zeros(size(tmp(:))),-tmp(:),size(tmp),...
zeros(size(tmp(:))),tmp(:),-0.5,round(size(tmp)./2));
%
% Smooth and threshold again.
%
spm_smooth(reshape(cctmp,size(tmp)),tmp,[4 4 4]);
cctmp = double(tmp(:) > 0.1);
%
% Next do a series of dilates followed by erodes
% to see if we can fill out the "holes" in the
% brain.
%
cctmp = dilate(cctmp,size(tmp));
cctmp = dilate(cctmp,size(tmp));
cctmp = dilate(cctmp,size(tmp));
cctmp = dilate(cctmp,size(tmp));
cctmp = erode(cctmp,size(tmp));
cctmp = erode(cctmp,size(tmp));
cctmp = erode(cctmp,size(tmp));
cctmp = erode(cctmp,size(tmp));
%
% Then dilate a little bit extra to make sure
% we have some margin around the brain.
%
cctmp = dilate(cctmp,size(tmp));
cctmp = dilate(cctmp,size(tmp));
%
% Smooth a little bit so that it is not
% all-or-nothing.
%
spm_smooth(reshape(cctmp,size(tmp)),tmp,[3 3 3]);
cctmp = tmp;
%
% Finally cut out relevant section
%
msk = cctmp(11:90,11:73,:);
oP.fname = '/d1/data/Petra/template/brainmask.img';
oP.dim = [dim P.dim(4)];
oP.mat = M2;
oP.pinfo = P.pinfo;
oP.descrip = 'T2 Rat Brain-mask in Paxonos Space';
oP = spm_write_vol(oP,msk);
%
% I have checked this brainmask, and it looks good.
% Jesper 14/6-02.
%
%
% Here I will shift the y- and z-axis such that
% the aspects of the different views are more
% like that of the human brain. This will mean
% that we can use the space in the SPM graphics
% window in a much more "economic way".
%
%
% Canonical
%
P = spm_vol('/d1/data/Petra/template/canonical_T2.img');
f = spm_read_vols(P);
of = zeros(size(f,1),size(f,3),size(f,2));
for i=1:size(f,1)
of(i,:,:) = squeeze(f(i,:,:))';
end
oP = P;
tmp = oP.dim(2);
oP.dim(2) = oP.dim(3);
oP.dim(3) = tmp;
tmp = oP.mat(2,4);
oP.mat(2,4) = oP.mat(3,4);
oP.mat(3,4) = tmp;
spm_write_vol(oP,of);
%
% Template
%
P = spm_vol('/d1/data/Petra/template/template_T2.img');
f = spm_read_vols(P);
of = zeros(size(f,1),size(f,3),size(f,2));
for i=1:size(f,1)
of(i,:,:) = squeeze(f(i,:,:))';
end
oP = P;
tmp = oP.dim(2);
oP.dim(2) = oP.dim(3);
oP.dim(3) = tmp;
tmp = oP.mat(2,4);
oP.mat(2,4) = oP.mat(3,4);
oP.mat(3,4) = tmp;
spm_write_vol(oP,of);
%
% Brainmask
%
P = spm_vol('/d1/data/Petra/template/brainmask.img');
f = spm_read_vols(P);
of = zeros(size(f,1),size(f,3),size(f,2));
for i=1:size(f,1)
of(i,:,:) = squeeze(f(i,:,:))';
end
oP = P;
tmp = oP.dim(2);
oP.dim(2) = oP.dim(3);
oP.dim(3) = tmp;
tmp = oP.mat(2,4);
oP.mat(2,4) = oP.mat(3,4);
oP.mat(3,4) = tmp;
spm_write_vol(oP,of);
%
% Here starts code pertaining to step 3,
% the creation of an "outer Paxinos contour".
%
axial = imread('/d1/data/Petra/rat_axial.jpg');
sagittal = imread('/d1/data/Petra/rat_sagittal.jpg');
coronal = imread('/d1/data/Petra/Petra.jpg');
%
% It seems that the "coronal" image has been created in a
% different way from the other two. Its matrix size is
% considerably larger, 9360x6800 compared to e.g. 494x715
% for the axial. We will therefor start out by down-sampling
% it.
%
coronal = coronal(1:5:end,1:5:end);
paxinos(1).ima = axial;
paxinos(1).slice = 'axial';
paxinos(1).bregma_level = -3.80;
paxinos(1).ia_level = 5.20;
paxinos(2).ima = sagittal;
paxinos(2).slice = 'sagittal';
paxinos(2).lateral_level = 0.40;
paxinos(3).ima = coronal;
paxinos(3).slice = 'coronal';
paxinos(3).bregma_level = -3.60;
paxinos(3).ia_level = 6.40;
clear axial sagittal coronal;
%
% What we need to do, for each image, is to create a train
% of polygons for the outer contour. This train should
% ideally be in "Paxonos co-ordinates", i.e. we need to
% establish an x and y scalefactor for each image, and
% also calibrate the zero-point.
%
figure('Position',[050 100 1.5*700 1.5*500])
imagesc(paxinos(1).ima); axis('image');
%
% First trace left y-axis from bottom to top.
%
[paxinos(1).lyax.x,paxinos(1).lyax.y] = ginput(12);
%
% Bottom x-axis from left to right.
%
[paxinos(1).bxax.x,paxinos(1).bxax.y] = ginput(17);
%
% And brain contour.
%
[paxinos(1).bc.x,paxinos(1).bc.y] = ginput;
save /d1/data/Petra/paxinos_cont.mat paxinos
%
% Find geometrical calibration (in LS sense)
% for axial slice.
%
X = [ones(17,1) paxinos(1).bxax.x];
y = [-8:8]';
b = pinv(X)*y;
paxinos(1).xb = b;
X = [ones(12,1) paxinos(1).lyax.y];
y = [-11:0]';
b = pinv(X)*y;
paxinos(1).yb = b;
paxinos(1).bc_calib.x = [ones(length(paxinos(1).bc.x),1) paxinos(1).bc.x] *...
paxinos(1).xb;
paxinos(1).bc_calib.y = [ones(length(paxinos(1).bc.y),1) paxinos(1).bc.y] *...
paxinos(1).yb;
%
% Sagittal
%
figure('Position',[050 100 5*220 5*160])
imagesc(paxinos(2).ima); axis('image');
%
% First trace left y-axis from bottom to top.
%
[paxinos(2).lyax.x,paxinos(2).lyax.y] = ginput(14);
%
% Bottom x-axis from left to right.
%
[paxinos(2).bxax.x,paxinos(2).bxax.y] = ginput(23);
%
% And brain contour.
%
[paxinos(2).bc.x,paxinos(2).bc.y] = ginput;
save /d1/data/Petra/paxinos_cont.mat paxinos
%
% Find geometrical calibration (in LS sense)
% for axial slice.
%
X = [ones(23,1) paxinos(2).bxax.x];
y = [6:-1:-16]';
b = pinv(X)*y;
paxinos(2).xb = b;
X = [ones(14,1) paxinos(2).lyax.y];
y = [-12:1]';
b = pinv(X)*y;
paxinos(2).yb = b;
paxinos(2).bc_calib.x = [ones(length(paxinos(2).bc.x),1) paxinos(2).bc.x] *...
paxinos(2).xb;
paxinos(2).bc_calib.y = [ones(length(paxinos(2).bc.y),1) paxinos(2).bc.y] *...
paxinos(2).yb;
%
% Coronal
%
figure('Position',[050 100 5*220 5*130])
imagesc(rot90(paxinos(3).ima)); axis('image');
%
% First trace left y-axis from bottom to top.
%
[paxinos(3).lyax.x,paxinos(3).lyax.y] = ginput(9);
[tmpx,tmpy] = ginput(8);
paxinos(3).lyax.x = [paxinos(3).lyax.x; tmpx];
paxinos(3).lyax.y = [paxinos(3).lyax.y; tmpy];
%
% Bottom x-axis from left to right.
%
[paxinos(3).bxax.x,paxinos(3).bxax.y] = ginput(12);
[tmpx,tmpy] = ginput(10);
paxinos(3).bxax.x = [paxinos(3).bxax.x; tmpx];
paxinos(3).bxax.y = [paxinos(3).bxax.y; tmpy];
%
% And brain contour.
%
[paxinos(3).bc.x,paxinos(3).bc.y] = ginput;
save /d1/data/Petra/paxinos_cont.mat paxinos
%
% Find geometrical calibration (in LS sense)
% for Coronal slice. There is a tiny bit of
% rotation on the scanned Coronal image, but
% I judge that as insignificant.
%
X = [ones(22,1) paxinos(3).bxax.x];
y = [6:-1:-15]';
b = pinv(X)*y;
paxinos(3).xb = b;
X = [ones(17,1) paxinos(3).lyax.y];
y = [-8:8]';
b = pinv(X)*y;
paxinos(3).yb = b;
paxinos(3).bc_calib.x = [ones(length(paxinos(3).bc.x),1) paxinos(3).bc.x] *...
paxinos(3).xb;
paxinos(3).bc_calib.y = [ones(length(paxinos(3).bc.y),1) paxinos(3).bc.y] *...
paxinos(3).yb;
save /d1/data/Petra/paxinos_cont.mat paxinos
%
% Now, lets try and create our own mip.mat
%
% It should be of size 400 wide and 364 high.
% Sagittal slice should be at coordinates (from upper left corner)
% ul=[6 26], lr=[225 155]
% Coronal
% ul=[6 180], lr=[225 340]
% Transversal
% ul=[236 26], lr[395 155]
figure('Position',[050 100 400 364])
%
% Sagittal
%
sag = axes('Position',[5/400 (364-25-130)/364 220/400 130/364]);
plot(paxinos(2).bc_calib.x,paxinos(2).bc_calib.y,'k','LineWidth',.5);
axis([-16 6 -12 1]);
set(sag,'XTickLabel',{}); set(sag,'YTickLabel',{});
set(sag,'XTick',[-14 -12 -10 -8 -6 -4 -2 2 4]); set(sag,'XGrid','On');
set(sag,'YTick',[-10 -8 -6 -4 -2]); set(sag,'YGrid','On');
hold on;
plot([-16 6],[0 0],'k','LineWidth',.5); plot([0 0],[-12 1],'k','LineWidth',.5);
hold off;
%
% Coronal
%
cor = axes('Position',[5/400 (364-25-130-24-160)/364 220/400 160/364]);
plot(paxinos(3).bc_calib.x,paxinos(3).bc_calib.y,'k','LineWidth',.5);
axis([-16 6 -8 8]);
set(cor,'XTickLabel',{}); set(cor,'YTickLabel',{});
set(cor,'XTick',[-14 -12 -10 -8 -6 -4 -2 2 4]); set(cor,'XGrid','On');
set(cor,'YTick',[-6 -4 -2 2 4 6]); set(cor,'YGrid','On');
hold on;
plot([-16 6],[0 0],'k','LineWidth',.5); plot([0 0],[-8 8],'k','LineWidth',.5);
hold off;
%
% Transversal
%
tra = axes('Position',[(5+220+10)/400 (364-25-130)/364 160/400 130/364]);
plot(paxinos(1).bc_calib.x,paxinos(1).bc_calib.y,'k','LineWidth',.5);
axis([-8 8 -12 1]);
set(tra,'XTickLabel',{}); set(tra,'YTickLabel',{});
set(tra,'XTick',[-6 -4 -2 2 4 6]); set(tra,'XGrid','On');
set(tra,'YTick',[-10 -8 -6 -4 -2]); set(tra,'YGrid','On');
hold on;
plot([-8 8],[0 0],'k','LineWidth',.5); plot([0 0],[-12 1],'k','LineWidth',.5);
hold off;
%
% Save this as .bmp file, read it back as image
% subsample it and write it back as .mat file.
%
set(gcf,'PaperPositionMode','auto');
saveas(gcf,'/d1/data/Petra/test','bmp');
ima = imread('/d1/data/Petra/test.bmp');
ima = double(~(ima(:,:,1)>0 | ima(:,:,2)>0 | ima(:,:,3)>0));
mip = zeros(364,400);
for i=1:364
for j=1:400
mip(i,j) = max(max(ima((i-1)*2+1:i*2,(j-1)*2+1:j*2)));
end
end
mip96 = rot90(mip,-1); % Compatible with MIP.mat
save /d1/data/Petra/MIP.mat mip96
%
% Now, determine suitable defaults for spatial normalisation
% (in spm_defaults) that would hopefully enable us to
% spatially normalise most scans without having to set
% origins and stuff manually for each rat.
% Notably, the zooming by a factor of 10 (remember, we
% work in tens of mm) and the swap of the y- and
% z-axes should be included in the defaults.
%
% Hence, the rotation (Pitch -pi/2 radians) and the
% zooms (10 10 -10) should be generally useful
% regardles of how the rat was positioned.
%
% In addition, the specific rat I have been working on
% was centered (in the up-down direction) in something
% which I suspect is the oesephagus. That is quite far
% away from the Bregma, so I will add an offset to
% compensate for that. This offset (115 tenths of mm)
% might be less general than the zooms and the rotation
% though (being specific to the positioning).
%
% Furthermore, in the nose-tail direction it has been
% centered on the "middle of the brain", which is a fair
% bit behind the Bregma. Hence I have added an offset
% also in that direction.
%
P = spm_vol('/d1/data/Petra/test.img');
f = spm_read_vols(P);
M = spm_matrix([0 -40 -115 -pi/2 0 0 10 10 -10]);
oP = P;
oP.mat = M*oP.mat;
spm_write_vol(oP,f);
%
% The numbers above look fine, and will be used
% in spm_defaults.m for rats.
% |
interprétation code matlab
Répondre avec citation
Partager