forked from MPUmri/ERRM
-
Notifications
You must be signed in to change notification settings - Fork 0
/
e03_3_collectSimMap.m
664 lines (547 loc) · 30 KB
/
e03_3_collectSimMap.m
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
% Collects results from previous step and exports summary as csv
% Estimated runtime: 60 seconds
%% Pre-setup
clearvars
addpath('./mfiles')
%% Setup
% Was non-negative least-squares performed?
didNonNeg = false;
% Output directories
outDir = './data/mapDownsample';
outDirNN = './data/mapDownsample-NN';
%%
tic
load('./data/simMap.mat');
% The settings below show be consistent with e03_2_analyzeSimMap.m
TRes = 1:60; % Range of temporal resolutions that were simulated
listSigmaC = 0:0.01:0.05; % Range of noise levels
% Reference region parameters
ktRR = 0.07;
kepRR = 0.5;
veRR = ktRR/kepRR;
repF = 100; % Number of replications for each noise level
trueKt = repmat(trueKt(:),[repF 1]);
trueKep = repmat(trueKep(:),[repF 1]);
trueVe = repmat(trueVe(:),[repF 1]);
trueVp = repmat(trueVp(:),[repF 1]);
%% Vp cutoff - Not actually used
% A range of vp can be defined here so that statistics are only calculated
% for curves with specific vp values
% Settings to [-inf inf] will use all simulated vp values
vpCutoff = [-inf inf]; % Min/Max true vp
vpMask = trueVp(:)>=vpCutoff(1) & trueVp(:)<=vpCutoff(2);
trueKt(~vpMask)=[];
trueKep(~vpMask)=[];
trueVe(~vpMask)=[];
trueVp(~vpMask)=[];
%% Collect results
for j=1:length(listSigmaC)
for i=1:length(TRes)
curFile = ['Downsample-Noise-' num2str(j) '-TRes-' num2str(i) '.mat'];
load(fullfile(outDir, 'CERRM', curFile));
load(fullfile(outDir, 'ETM', curFile));
% Record number of imaginary estimates from ERRM
numImagE(i,j) = sum( sum(abs(imag(pkERRM')))>0 );
% Use only real part when calculating percent error for ERRM
pkERRM = real(pkERRM);
% OPTIONAL
% For ERRM: we could exclude fits where tissue of interest and
% reference tissue have same kep values, as the ERRM fails to fit those cases
goodERRM = (trueKep~=kepRR);
% However, we will not remove those fits, so that the same number of
% fits are used for all models
% (It doesn't really change the results anyways)
goodERRM(:) = true; % Comment this line to ignore voxels where kep=kepRR
%% Apply vpCutOff
% Does nothing if vpCutoff = [-inf inf] (Default setting)
pkCERRM(~vpMask,:)=[];
pkCLRRM(~vpMask,:)=[];
pkERRM(~vpMask,:)=[];
pkERTM(~vpMask,:)=[];
pkETM(~vpMask,:)=[];
pkLRRM(~vpMask,:)=[];
pkRTM(~vpMask,:)=[];
pkTM(~vpMask,:)=[];
%% Collect mean, stdDev, and quantiles of the percent errors
% Tofts Model - TM
[ktError, meanKtErrT(i,j), stdKtErrT(i,j)] = PercentError(pkTM(:,1),trueKt(:));
[veError, meanVeErrT(i,j), stdVeErrT(i,j)] = PercentError(pkTM(:,1)./pkTM(:,2),trueVe(:));
[kepError, meanKepErrT(i,j), stdKepErrT(i,j)] = PercentError(pkTM(:,2),trueKep(:));
qKtT(i,j,:) = quantile(ktError,[.05 .25 .5 .75 .95]);
qVeT(i,j,:) = quantile(veError,[.05 .25 .5 .75 .95]);
qKepT(i,j,:) = quantile(kepError,[.05 .25 .5 .75 .95]);
% Extended Tofts Model - ETM
[ktError, meanKtErrET(i,j), stdKtErrET(i,j)] = PercentError(pkETM(:,1),trueKt(:));
[veError, meanVeErrET(i,j), stdVeErrET(i,j)] = PercentError(pkETM(:,1)./pkETM(:,2),trueVe(:));
[kepError, meanKepErrET(i,j), stdKepErrET(i,j)] = PercentError(pkETM(:,2),trueKep(:));
[vpError, meanVpErrET(i,j), stdVpErrET(i,j)] = PercentError(pkETM(:,3),trueVp(:));
qKtET(i,j,:) = quantile(ktError,[.05 .25 .5 .75 .95]);
qVeET(i,j,:) = quantile(veError,[.05 .25 .5 .75 .95]);
qKepET(i,j,:) = quantile(kepError,[.05 .25 .5 .75 .95]);
qVpET(i,j,:) = quantile(vpError,[.05 .25 .5 .75 .95]);
% Reference Tissue Method - RTM (not mentioned in manuscript)
[ktError, meanKtErrRT(i,j), stdKtErrRT(i,j)] = PercentError(pkRTM(:,1),trueKt(:));
[veError, meanVeErrRT(i,j), stdVeErrRT(i,j)] = PercentError(pkRTM(:,1)./pkRTM(:,2),trueVe(:));
[kepError, meanKepErrRT(i,j), stdKepErrRT(i,j)] = PercentError(pkRTM(:,2),trueKep(:));
qKtRT(i,j,:) = quantile(ktError,[.05 .25 .5 .75 .95]);
qVeRT(i,j,:) = quantile(veError,[.05 .25 .5 .75 .95]);
qKepRT(i,j,:) = quantile(kepError,[.05 .25 .5 .75 .95]);
% Extended Reference Tissue Method - ERTM (not mentioned in manuscript)
[ktError, meanKtErrERT(i,j), stdKtErrERT(i,j)] = PercentError(pkERTM(:,1),trueKt(:));
[veError, meanVeErrERT(i,j), stdVeErrERT(i,j)] = PercentError(pkERTM(:,1)./pkERTM(:,2),trueVe(:));
[kepError, meanKepErrERT(i,j), stdKepErrERT(i,j)] = PercentError(pkERTM(:,2),trueKep(:));
[vpError, meanVpErrERT(i,j), stdVpErrERT(i,j)] = PercentError(pkERTM(:,3),trueVp(:));
qKtERT(i,j,:) = quantile(ktError,[.05 .25 .5 .75 .95]);
qVeERT(i,j,:) = quantile(veError,[.05 .25 .5 .75 .95]);
qKepERT(i,j,:) = quantile(kepError,[.05 .25 .5 .75 .95]);
qVpERT(i,j,:) = quantile(vpError,[.05 .25 .5 .75 .95]);
% Constrained Extended Reference Region Model - CERRM
[ktError, meanKtErrCE(i,j), stdKtErrCE(i,j), badValKtCE(i,j)] = PercentError(ktRR*pkCERRM(:,1),trueKt(:));
[veError, meanVeErrCE(i,j), stdVeErrCE(i,j), badValVeCE(i,j)] = PercentError(veRR*pkCERRM(:,2),trueVe(:));
[kepError, meanKepErrCE(i,j), stdKepErrCE(i,j), badValKepCE(i,j)] = PercentError(pkCERRM(:,3),trueKep(:));
[vpError, meanVpErrCE(i,j), stdVpErrCE(i,j), badValVpCE(i,j)] = PercentError(ktRR*pkCERRM(:,4),trueVp(:));
qKtCE(i,j,:) = quantile(ktError,[.05 .25 .5 .75 .95]);
qVeCE(i,j,:) = quantile(veError,[.05 .25 .5 .75 .95]);
qKepCE(i,j,:) = quantile(kepError,[.05 .25 .5 .75 .95]);
qVpCE(i,j,:) = quantile(vpError,[.05 .25 .5 .75 .95]);
% Extended Reference Region Model - ERRM
[ktError, meanKtErrE(i,j), stdKtErrE(i,j), badValKtE(i,j)] = PercentError(ktRR*pkERRM(:,1),trueKt(:));
[veError, meanVeErrE(i,j), stdVeErrE(i,j), badValVeE(i,j)] = PercentError(veRR*pkERRM(:,2),trueVe(:));
[kepError, meanKepErrE(i,j), stdKepErrE(i,j), badValKepE(i,j)] = PercentError(pkERRM(:,3),trueKep(:));
[vpError, meanVpErrE(i,j), stdVpErrE(i,j), badValVpE(i,j)] = PercentError(ktRR*pkERRM(:,4),trueVp(:));
qKtE(i,j,:) = quantile(ktError(goodERRM),[.05 .25 .5 .75 .95]);
qVeE(i,j,:) = quantile(veError(goodERRM),[.05 .25 .5 .75 .95]);
qKepE(i,j,:) = quantile(kepError(goodERRM),[.05 .25 .5 .75 .95]);
qVpE(i,j,:) = quantile(vpError,[.05 .25 .5 .75 .95]);
% Reference Region Model - RRM
[ktError, meanKtErrR(i,j), stdKtErrR(i,j)] = PercentError(ktRR*pkLRRM(:,1),trueKt(:));
[veError, meanVeErrR(i,j), stdVeErrR(i,j)] = PercentError(veRR*pkLRRM(:,2),trueVe(:));
[kepError, meanKepErrR(i,j), stdKepErrR(i,j)] = PercentError(pkLRRM(:,3),trueKep(:));
qKtR(i,j,:) = quantile(ktError,[.05 .25 .5 .75 .95]);
qVeR(i,j,:) = quantile(veError,[.05 .25 .5 .75 .95]);
qKepR(i,j,:) = quantile(kepError,[.05 .25 .5 .75 .95]);
% Non-extended Contrained Reference Region Model - CRRM (not mentioned in manuscript)
[ktError, meanKtErrCR(i,j), stdKtErrCR(i,j)] = PercentError(ktRR*pkCLRRM(:,1),trueKt(:));
[veError, meanVeErrCR(i,j), stdVeErrCR(i,j)] = PercentError(veRR*pkCLRRM(:,2),trueVe(:));
[kepError, meanKepErrCR(i,j), stdKepErrCR(i,j)] = PercentError(pkCLRRM(:,3),trueKep(:));
qKtCR(i,j,:) = quantile(ktError,[.05 .25 .5 .75 .95]);
qVeCR(i,j,:) = quantile(veError,[.05 .25 .5 .75 .95]);
qKepCR(i,j,:) = quantile(kepError,[.05 .25 .5 .75 .95]);
%% Load fits using NNLS - if the fits were performed
% This is pretty much a copy/paste job of the code above, with minor
% changes. Minimal comments ahead.
if didNonNeg
load(fullfile(outDirNN, 'CERRM', curFile));
load(fullfile(outDirNN, 'ETM', curFile));
numImagEnn(i,j) = sum( sum(abs(imag(pkERRM')))>0 );
pkERRM = real(pkERRM);
% Apply vpCutOff
pkCERRM(~vpMask,:)=[];
pkERRM(~vpMask,:)=[];
pkERTM(~vpMask,:)=[];
pkETM(~vpMask,:)=[];
pkRTM(~vpMask,:)=[];
pkTM(~vpMask,:)=[];
% TM
[ktError, meanKtErrTnn(i,j), stdKtErrTnn(i,j)] = PercentError(pkTM(:,1),trueKt(:));
[veError, meanVeErrTnn(i,j), stdVeErrTnn(i,j)] = PercentError(pkTM(:,1)./pkTM(:,2),trueVe(:));
[kepError, meanKepErrTnn(i,j), stdKepErrTnn(i,j)] = PercentError(pkTM(:,2),trueKep(:));
qKtTnn(i,j,:) = quantile(ktError,[.05 .25 .5 .75 .95]);
qVeTnn(i,j,:) = quantile(veError,[.05 .25 .5 .75 .95]);
qKepTnn(i,j,:) = quantile(kepError,[.05 .25 .5 .75 .95]);
% ETM
[ktError, meanKtErrETnn(i,j), stdKtErrETnn(i,j)] = PercentError(pkETM(:,1),trueKt(:));
[veError, meanVeErrETnn(i,j), stdVeErrETnn(i,j)] = PercentError(pkETM(:,1)./pkETM(:,2),trueVe(:));
[kepError, meanKepErrETnn(i,j), stdKepErrETnn(i,j)] = PercentError(pkETM(:,2),trueKep(:));
[vpError, meanVpErrETnn(i,j), stdVpErrETnn(i,j)] = PercentError(pkETM(:,3),trueVp(:));
qKtETnn(i,j,:) = quantile(ktError,[.05 .25 .5 .75 .95]);
qVeETnn(i,j,:) = quantile(veError,[.05 .25 .5 .75 .95]);
qKepETnn(i,j,:) = quantile(kepError,[.05 .25 .5 .75 .95]);
qVpETnn(i,j,:) = quantile(vpError,[.05 .25 .5 .75 .95]);
% RTM
[ktError, meanKtErrRTnn(i,j), stdKtErrRTnn(i,j)] = PercentError(pkRTM(:,1),trueKt(:));
[veError, meanVeErrRTnn(i,j), stdVeErrRTnn(i,j)] = PercentError(pkRTM(:,1)./pkRTM(:,2),trueVe(:));
[kepError, meanKepErrRTnn(i,j), stdKepErrRTnn(i,j)] = PercentError(pkRTM(:,2),trueKep(:));
qKtRTnn(i,j,:) = quantile(ktError,[.05 .25 .5 .75 .95]);
qVeRTnn(i,j,:) = quantile(veError,[.05 .25 .5 .75 .95]);
qKepRTnn(i,j,:) = quantile(kepError,[.05 .25 .5 .75 .95]);
% ERTM
[ktError, meanKtErrERTnn(i,j), stdKtErrERTnn(i,j)] = PercentError(pkERTM(:,1),trueKt(:));
[veError, meanVeErrERTnn(i,j), stdVeErrERTnn(i,j)] = PercentError(pkERTM(:,1)./pkERTM(:,2),trueVe(:));
[kepError, meanKepErrERTnn(i,j), stdKepErrERTnn(i,j)] = PercentError(pkERTM(:,2),trueKep(:));
[vpError, meanVpErrERTnn(i,j), stdVpErrERTnn(i,j)] = PercentError(pkERTM(:,3),trueVp(:));
qKtERTnn(i,j,:) = quantile(ktError,[.05 .25 .5 .75 .95]);
qVeERTnn(i,j,:) = quantile(veError,[.05 .25 .5 .75 .95]);
qKepERTnn(i,j,:) = quantile(kepError,[.05 .25 .5 .75 .95]);
qVpERTnn(i,j,:) = quantile(vpError,[.05 .25 .5 .75 .95]);
% CERRM
[ktError, meanKtErrCEnn(i,j), stdKtErrCEnn(i,j), badValKtCEnn(i,j)] = PercentError(ktRR*pkCERRM(:,1),trueKt(:));
[veError, meanVeErrCEnn(i,j), stdVeErrCEnn(i,j), badValVeCEnn(i,j)] = PercentError(veRR*pkCERRM(:,2),trueVe(:));
[kepError, meanKepErrCEnn(i,j), stdKepErrCEnn(i,j), badValKepCEnn(i,j)] = PercentError(pkCERRM(:,3),trueKep(:));
[vpError, meanVpErrCEnn(i,j), stdVpErrCEnn(i,j), badValVpCEnn(i,j)] = PercentError(ktRR*pkCERRM(:,4),trueVp(:));
qKtCEnn(i,j,:) = quantile(ktError,[.05 .25 .5 .75 .95]);
qVeCEnn(i,j,:) = quantile(veError,[.05 .25 .5 .75 .95]);
qKepCEnn(i,j,:) = quantile(kepError,[.05 .25 .5 .75 .95]);
qVpCEnn(i,j,:) = quantile(vpError,[.05 .25 .5 .75 .95]);
% ERRM
[ktError, meanKtErrEnn(i,j), stdKtErrEnn(i,j), badValKtEnn(i,j)] = PercentError(ktRR*pkERRM(:,1),trueKt(:));
[veError, meanVeErrEnn(i,j), stdVeErrEnn(i,j), badValVeEnn(i,j)] = PercentError(veRR*pkERRM(:,2),trueVe(:));
[kepError, meanKepErrEnn(i,j), stdKepErrEnn(i,j), badValKepEnn(i,j)] = PercentError(pkERRM(:,3),trueKep(:));
[vpError, meanVpErrEnn(i,j), stdVpErrEnn(i,j), badValVpEnn(i,j)] = PercentError(ktRR*pkERRM(:,4),trueVp(:));
qKtEnn(i,j,:) = quantile(ktError(goodERRM),[.05 .25 .5 .75 .95]);
qVeEnn(i,j,:) = quantile(veError(goodERRM),[.05 .25 .5 .75 .95]);
qKepEnn(i,j,:) = quantile(kepError(goodERRM),[.05 .25 .5 .75 .95]);
qVpEnn(i,j,:) = quantile(vpError,[.05 .25 .5 .75 .95]);
end
end
end
%% Export mean and std of percent errors as csv
outFile = fullfile('./dataResults','e03a-downsampleMapResultsMean.csv');
hdr='FitMethod,TemporalRes,sigmaC,errKt,errVe,errKep,errVp,stdKt,stdVe,stdKep,stdVp';
outID = fopen(outFile, 'w+');
fprintf(outID, '%s\n', hdr); % Print header into csv file
for j=1:length(listSigmaC)
for i=1:length(TRes)
outLine = {'ETM',TRes(i),listSigmaC(j),meanKtErrET(i,j),meanVeErrET(i,j),meanKepErrET(i,j),meanVpErrET(i,j),...
stdKtErrET(i,j),stdVeErrET(i,j),stdKepErrET(i,j),stdVpErrET(i,j)};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'ERTM',TRes(i),listSigmaC(j),meanKtErrERT(i,j),meanVeErrERT(i,j),meanKepErrERT(i,j),meanVpErrERT(i,j),...
stdKtErrERT(i,j),stdVeErrERT(i,j),stdKepErrERT(i,j),stdVpErrERT(i,j)};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'RTM',TRes(i),listSigmaC(j),meanKtErrRT(i,j),meanVeErrRT(i,j),meanKepErrRT(i,j),NaN,...
stdKtErrRT(i,j),stdVeErrRT(i,j),stdKepErrRT(i,j),NaN};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'ERRM',TRes(i),listSigmaC(j),meanKtErrE(i,j),meanVeErrE(i,j),meanKepErrE(i,j),meanVpErrE(i,j),...
stdKtErrE(i,j),stdVeErrE(i,j),stdKepErrE(i,j),stdVpErrE(i,j)};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'CERRM',TRes(i),listSigmaC(j),meanKtErrCE(i,j),meanVeErrCE(i,j),meanKepErrCE(i,j),meanVpErrCE(i,j),...
stdKtErrCE(i,j),stdVeErrCE(i,j),stdKepErrCE(i,j),stdVpErrCE(i,j)};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'TM',TRes(i),listSigmaC(j),meanKtErrT(i,j),meanVeErrT(i,j),meanKepErrT(i,j),NaN,...
stdKtErrT(i,j),stdVeErrT(i,j),stdKepErrT(i,j),NaN};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'RRM',TRes(i),listSigmaC(j),meanKtErrR(i,j),meanVeErrR(i,j),meanKepErrR(i,j),NaN,...
stdKtErrR(i,j),stdVeErrR(i,j),stdKepErrR(i,j),NaN};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'CRRM',TRes(i),listSigmaC(j),meanKtErrCR(i,j),meanVeErrCR(i,j),meanKepErrCR(i,j),NaN,...
stdKtErrCR(i,j),stdVeErrCR(i,j),stdKepErrCR(i,j),NaN};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
% Export results from NNLS fits, if they were performed
if ~didNonNeg
continue
end
outLine = {'ETM-NN',TRes(i),listSigmaC(j),meanKtErrETnn(i,j),meanVeErrETnn(i,j),meanKepErrETnn(i,j),meanVpErrETnn(i,j),...
stdKtErrETnn(i,j),stdVeErrETnn(i,j),stdKepErrETnn(i,j),stdVpErrETnn(i,j)};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'ERTM-NN',TRes(i),listSigmaC(j),meanKtErrERTnn(i,j),meanVeErrERTnn(i,j),meanKepErrERTnn(i,j),meanVpErrERTnn(i,j),...
stdKtErrERTnn(i,j),stdVeErrERTnn(i,j),stdKepErrERTnn(i,j),stdVpErrERTnn(i,j)};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'RTM-NN',TRes(i),listSigmaC(j),meanKtErrRTnn(i,j),meanVeErrRTnn(i,j),meanKepErrRTnn(i,j),NaN,...
stdKtErrRTnn(i,j),stdVeErrRTnn(i,j),stdKepErrRTnn(i,j),NaN};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'ERRM-NN',TRes(i),listSigmaC(j),meanKtErrEnn(i,j),meanVeErrEnn(i,j),meanKepErrEnn(i,j),meanVpErrEnn(i,j),...
stdKtErrEnn(i,j),stdVeErrEnn(i,j),stdKepErrEnn(i,j),stdVpErrEnn(i,j)};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'CERRM-NN',TRes(i),listSigmaC(j),meanKtErrCEnn(i,j),meanVeErrCEnn(i,j),meanKepErrCEnn(i,j),meanVpErrCEnn(i,j),...
stdKtErrCEnn(i,j),stdVeErrCEnn(i,j),stdKepErrCEnn(i,j),stdVpErrCEnn(i,j)};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'TM-NN',TRes(i),listSigmaC(j),meanKtErrTnn(i,j),meanVeErrTnn(i,j),meanKepErrTnn(i,j),NaN,...
stdKtErrTnn(i,j),stdVeErrTnn(i,j),stdKepErrTnn(i,j),NaN};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
end
end
fclose(outID);
%% Export quartiles of percent error as csv
outFile = fullfile('./dataResults',['e03b-downsampleMapResultsQuantiles.csv']);
hdr=['FitMethod,TemporalRes,sigmaC,q5Kt,q25Kt,q50Kt,q75Kt,q95Kt,q5Ve,q25Ve,q50Ve,q75Ve,q95Ve,q5Kep,q25Kep,q50Kep,q75Kep,q95Kep,q5Vp,q25Vp,q50Vp,q75Vp,q95Vp'];
outID = fopen(outFile, 'w+');
fprintf(outID, '%s\n', hdr); % Print header into csv file
for j=1:length(listSigmaC)
for i=1:length(TRes)
outLine = {'ETM',TRes(i),listSigmaC(j),...
qKtET(i,j,1),qKtET(i,j,2),qKtET(i,j,3),qKtET(i,j,4),qKtET(i,j,5),...
qVeET(i,j,1),qVeET(i,j,2),qVeET(i,j,3),qVeET(i,j,4),qVeET(i,j,5),...
qKepET(i,j,1),qKepET(i,j,2),qKepET(i,j,3),qKepET(i,j,4),qKepET(i,j,5),...
qVpET(i,j,1),qVpET(i,j,2),qVpET(i,j,3),qVpET(i,j,4),qVpET(i,j,5),...
};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'ERTM',TRes(i),listSigmaC(j),...
qKtERT(i,j,1),qKtERT(i,j,2),qKtERT(i,j,3),qKtERT(i,j,4),qKtERT(i,j,5),...
qVeERT(i,j,1),qVeERT(i,j,2),qVeERT(i,j,3),qVeERT(i,j,4),qVeERT(i,j,5),...
qKepERT(i,j,1),qKepERT(i,j,2),qKepERT(i,j,3),qKepERT(i,j,4),qKepERT(i,j,5),...
qVpERT(i,j,1),qVpERT(i,j,2),qVpERT(i,j,3),qVpERT(i,j,4),qVpERT(i,j,5),...
};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'RTM',TRes(i),listSigmaC(j),...
qKtRT(i,j,1),qKtRT(i,j,2),qKtRT(i,j,3),qKtRT(i,j,4),qKtRT(i,j,5),...
qVeRT(i,j,1),qVeRT(i,j,2),qVeRT(i,j,3),qVeRT(i,j,4),qVeRT(i,j,5),...
qKepRT(i,j,1),qKepRT(i,j,2),qKepRT(i,j,3),qKepRT(i,j,4),qKepRT(i,j,5),...
NaN,NaN,NaN,NaN,NaN...
};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'CERRM',TRes(i),listSigmaC(j),...
qKtCE(i,j,1),qKtCE(i,j,2),qKtCE(i,j,3),qKtCE(i,j,4),qKtCE(i,j,5),...
qVeCE(i,j,1),qVeCE(i,j,2),qVeCE(i,j,3),qVeCE(i,j,4),qVeCE(i,j,5),...
qKepCE(i,j,1),qKepCE(i,j,2),qKepCE(i,j,3),qKepCE(i,j,4),qKepCE(i,j,5),...
qVpCE(i,j,1),qVpCE(i,j,2),qVpCE(i,j,3),qVpCE(i,j,4),qVpCE(i,j,5),...
};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'ERRM',TRes(i),listSigmaC(j),...
qKtE(i,j,1),qKtE(i,j,2),qKtE(i,j,3),qKtE(i,j,4),qKtE(i,j,5),...
qVeE(i,j,1),qVeE(i,j,2),qVeE(i,j,3),qVeE(i,j,4),qVeE(i,j,5),...
qKepE(i,j,1),qKepE(i,j,2),qKepE(i,j,3),qKepE(i,j,4),qKepE(i,j,5),...
qVpE(i,j,1),qVpE(i,j,2),qVpE(i,j,3),qVpE(i,j,4),qVpE(i,j,5),...
};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'TM',TRes(i),listSigmaC(j),...
qKtT(i,j,1),qKtT(i,j,2),qKtT(i,j,3),qKtT(i,j,4),qKtT(i,j,5),...
qVeT(i,j,1),qVeT(i,j,2),qVeT(i,j,3),qVeT(i,j,4),qVeT(i,j,5),...
qKepT(i,j,1),qKepT(i,j,2),qKepT(i,j,3),qKepT(i,j,4),qKepT(i,j,5),...
NaN,NaN,NaN,NaN,NaN,...
};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'RRM',TRes(i),listSigmaC(j),...
qKtR(i,j,1),qKtR(i,j,2),qKtR(i,j,3),qKtR(i,j,4),qKtR(i,j,5),...
qVeR(i,j,1),qVeR(i,j,2),qVeR(i,j,3),qVeR(i,j,4),qVeR(i,j,5),...
qKepR(i,j,1),qKepR(i,j,2),qKepR(i,j,3),qKepR(i,j,4),qKepR(i,j,5),...
NaN,NaN,NaN,NaN,NaN,...
};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'CRRM',TRes(i),listSigmaC(j),...
qKtCR(i,j,1),qKtCR(i,j,2),qKtCR(i,j,3),qKtCR(i,j,4),qKtCR(i,j,5),...
qVeCR(i,j,1),qVeCR(i,j,2),qVeCR(i,j,3),qVeCR(i,j,4),qVeCR(i,j,5),...
qKepCR(i,j,1),qKepCR(i,j,2),qKepCR(i,j,3),qKepCR(i,j,4),qKepCR(i,j,5),...
NaN,NaN,NaN,NaN,NaN,...
};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
% Export results from NNLS fits, if they were performed
if ~didNonNeg
continue
end
outLine = {'ETM-NN',TRes(i),listSigmaC(j),...
qKtETnn(i,j,1),qKtETnn(i,j,2),qKtETnn(i,j,3),qKtETnn(i,j,4),qKtETnn(i,j,5),...
qVeETnn(i,j,1),qVeETnn(i,j,2),qVeETnn(i,j,3),qVeETnn(i,j,4),qVeETnn(i,j,5),...
qKepETnn(i,j,1),qKepETnn(i,j,2),qKepETnn(i,j,3),qKepETnn(i,j,4),qKepETnn(i,j,5),...
qVpETnn(i,j,1),qVpETnn(i,j,2),qVpETnn(i,j,3),qVpETnn(i,j,4),qVpETnn(i,j,5),...
};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'ERTM-NN',TRes(i),listSigmaC(j),...
qKtERTnn(i,j,1),qKtERTnn(i,j,2),qKtERTnn(i,j,3),qKtERTnn(i,j,4),qKtERTnn(i,j,5),...
qVeERTnn(i,j,1),qVeERTnn(i,j,2),qVeERTnn(i,j,3),qVeERTnn(i,j,4),qVeERTnn(i,j,5),...
qKepERTnn(i,j,1),qKepERTnn(i,j,2),qKepERTnn(i,j,3),qKepERTnn(i,j,4),qKepERTnn(i,j,5),...
qVpERTnn(i,j,1),qVpERTnn(i,j,2),qVpERTnn(i,j,3),qVpERTnn(i,j,4),qVpERTnn(i,j,5),...
};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'RTM-NN',TRes(i),listSigmaC(j),...
qKtRTnn(i,j,1),qKtRTnn(i,j,2),qKtRTnn(i,j,3),qKtRTnn(i,j,4),qKtRTnn(i,j,5),...
qVeRTnn(i,j,1),qVeRTnn(i,j,2),qVeRTnn(i,j,3),qVeRTnn(i,j,4),qVeRTnn(i,j,5),...
qKepRTnn(i,j,1),qKepRTnn(i,j,2),qKepRTnn(i,j,3),qKepRTnn(i,j,4),qKepRTnn(i,j,5),...
NaN,NaN,NaN,NaN,NaN...
};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'CERRM-NN',TRes(i),listSigmaC(j),...
qKtCEnn(i,j,1),qKtCEnn(i,j,2),qKtCEnn(i,j,3),qKtCEnn(i,j,4),qKtCEnn(i,j,5),...
qVeCEnn(i,j,1),qVeCEnn(i,j,2),qVeCEnn(i,j,3),qVeCEnn(i,j,4),qVeCEnn(i,j,5),...
qKepCEnn(i,j,1),qKepCEnn(i,j,2),qKepCEnn(i,j,3),qKepCEnn(i,j,4),qKepCEnn(i,j,5),...
qVpCEnn(i,j,1),qVpCEnn(i,j,2),qVpCEnn(i,j,3),qVpCEnn(i,j,4),qVpCEnn(i,j,5),...
};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'ERRM-NN',TRes(i),listSigmaC(j),...
qKtEnn(i,j,1),qKtEnn(i,j,2),qKtEnn(i,j,3),qKtEnn(i,j,4),qKtEnn(i,j,5),...
qVeEnn(i,j,1),qVeEnn(i,j,2),qVeEnn(i,j,3),qVeEnn(i,j,4),qVeEnn(i,j,5),...
qKepEnn(i,j,1),qKepEnn(i,j,2),qKepEnn(i,j,3),qKepEnn(i,j,4),qKepEnn(i,j,5),...
qVpEnn(i,j,1),qVpEnn(i,j,2),qVpEnn(i,j,3),qVpEnn(i,j,4),qVpEnn(i,j,5),...
};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
outLine = {'TM-NN',TRes(i),listSigmaC(j),...
qKtTnn(i,j,1),qKtTnn(i,j,2),qKtTnn(i,j,3),qKtTnn(i,j,4),qKtTnn(i,j,5),...
qVeTnn(i,j,1),qVeTnn(i,j,2),qVeTnn(i,j,3),qVeTnn(i,j,4),qVeTnn(i,j,5),...
qKepTnn(i,j,1),qKepTnn(i,j,2),qKepTnn(i,j,3),qKepTnn(i,j,4),qKepTnn(i,j,5),...
NaN,NaN,NaN,NaN,NaN,...
};
fprintf(outID,'%s,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f,%f\n', outLine{:});
end
end
fclose(outID);
toc
%% Plots
%% Median +/- interquartile range vs StdDev of Noise
myX = listSigmaC;
myTres = 5;
lErr = 2;
hErr = 4;
clrERRM = [247,151,86]./255;
clrCERRM = [126,47,142]./255;
figure('Position',[300 300 1200 500])
subplot(1,4,1)
errorbar(myX,qKtE(myTres,:,3),abs(qKtE(myTres,:,lErr)-qKtE(myTres,:,3)),abs(qKtE(myTres,:,hErr)-qKtE(myTres,:,3)),'linewidth',3,'linestyle','--','Color',clrERRM)
hold on
errorbar(myX,qKtCE(myTres,:,3),abs(qKtCE(myTres,:,lErr)-qKtCE(myTres,:,3)),abs(qKtCE(myTres,:,hErr)-qKtCE(myTres,:,3)),'linewidth',3,'Color',clrCERRM)
hold off
ylim([-70 10])
xlabel('StdDev [mM]')
ylabel('Percent Error')
title('Ktrans')
subplot(1,4,2)
errorbar(myX,qKepE(myTres,:,3),abs(qKepE(myTres,:,lErr)-qKepE(myTres,:,3)),abs(qKepE(myTres,:,hErr)-qKepE(myTres,:,3)),'linewidth',3,'linestyle','--','Color',clrERRM)
hold on
errorbar(myX,qKepCE(myTres,:,3),abs(qKepCE(myTres,:,lErr)-qKepCE(myTres,:,3)),abs(qKepCE(myTres,:,hErr)-qKepCE(myTres,:,3)),'linewidth',3,'Color',clrCERRM)
hold off
ylim([-50 20])
xlabel('StdDev [mM]')
ylabel('Percent Error')
title('kep')
subplot(1,4,3)
errorbar(myX,qVeE(myTres,:,3),abs(qVeE(myTres,:,lErr)-qVeE(myTres,:,3)),abs(qVeE(myTres,:,hErr)-qVeE(myTres,:,3)),'linewidth',3,'linestyle','--','Color',clrERRM)
hold on
errorbar(myX,qVeCE(myTres,:,3),abs(qVeCE(myTres,:,lErr)-qVeCE(myTres,:,3)),abs(qVeCE(myTres,:,hErr)-qVeCE(myTres,:,3)),'linewidth',3,'Color',clrCERRM)
hold off
ylim([-50 20])
xlabel('StdDev [mM]')
ylabel('Percent Error')
title('ve')
subplot(1,4,4)
errorbar(myX,qVpE(myTres,:,3),abs(qVpE(myTres,:,lErr)-qVpE(myTres,:,3)),abs(qVpE(myTres,:,hErr)-qVpE(myTres,:,3)),'linewidth',3,'linestyle','--','Color',clrERRM)
hold on
errorbar(myX,qVpCE(myTres,:,3),abs(qVpCE(myTres,:,lErr)-qVpCE(myTres,:,3)),abs(qVpCE(myTres,:,hErr)-qVpCE(myTres,:,3)),'linewidth',3,'Color',clrCERRM)
hold off
ylim([-40 160])
xlabel('StdDev [mM]')
ylabel('Percent Error')
title('vp')
legend('ERRM','CERRM')
%% Ratios of interquartile range between ERRM and CERRM
quartRatio_Kt=(qKtE(myTres,:,hErr)-qKtE(myTres,:,lErr))./(qKtCE(myTres,:,hErr)-qKtCE(myTres,:,lErr));
quartRatio_Kep=(qKepE(myTres,:,hErr)-qKepE(myTres,:,lErr))./(qKepCE(myTres,:,hErr)-qKepCE(myTres,:,lErr));
quartRatio_Ve=(qVeE(myTres,:,hErr)-qVeE(myTres,:,lErr))./(qVeCE(myTres,:,hErr)-qVeCE(myTres,:,lErr));
quartRatio_Vp=(qVpE(myTres,:,hErr)-qVpE(myTres,:,lErr))./(qVpCE(myTres,:,hErr)-qVpCE(myTres,:,lErr));
figure
boxplot([quartRatio_Kt; quartRatio_Kep; quartRatio_Ve; quartRatio_Vp]', ...
'Labels',{'Ktrans','kep','ve','vp'})
title('Ratio of Interquartile Range between ERRM and CERRM')
ylim([0 4])
% The above figure shows the ratio of the interquartile range, i.e. the error
% bars in Figure 2, between the ERRM and CERRM. For example, a value of 2
% means that the ERRM's error bars are twice as large as the CERRM.
% The boxplot shows the spread of this ratio for the different noise
% levels.
% The manuscript undersells the improvement of the CERRM by saying that its
% error bars were smaller by a factor of 1.5, 3, 1.5, and ~1 for Ktrans,
% kep, ve, and vp, respectively.
% The figure shows a median improvement by a factor of 2, 3.2, 2, and ~1,
% respectively.
%% Median +/- interquartile range vs Temporal Resolution
mySig = 2;
lErr = 2;
hErr = 4;
xRange = 1:30;
cCE = [.5 .2 .55];
cET = [.3 .7 .6];
figure('Position',[300 300 1200 500])
subplot(1,4,1)
hold on
shadedErrorPlot(TRes(xRange)',qKtET(xRange,mySig,3),qKtET(xRange,mySig,lErr),qKtET(xRange,mySig,hErr),cET);
shadedErrorPlot(TRes(xRange)',qKtCE(xRange,mySig,3),qKtCE(xRange,mySig,lErr),qKtCE(xRange,mySig,hErr),cCE);
hold off
ylim([-60 40])
xlabel('Temporal Resolution [s]')
ylabel('Percent Error')
title('Ktrans')
legend('','ETM','','CERRM')
subplot(1,4,2)
hold on
shadedErrorPlot(TRes(xRange)',qKepET(xRange,mySig,3),qKepET(xRange,mySig,lErr),qKepET(xRange,mySig,hErr),cET);
shadedErrorPlot(TRes(xRange)',qKepCE(xRange,mySig,3),qKepCE(xRange,mySig,lErr),qKepCE(xRange,mySig,hErr),cCE);
hold off
ylim([-60 40])
xlabel('Temporal Resolution [s]')
ylabel('Percent Error')
title('kep')
subplot(1,4,3)
hold on
shadedErrorPlot(TRes(xRange)',qVeET(xRange,mySig,3),qVeET(xRange,mySig,lErr),qVeET(xRange,mySig,hErr),cET);
shadedErrorPlot(TRes(xRange)',qVeCE(xRange,mySig,3),qVeCE(xRange,mySig,lErr),qVeCE(xRange,mySig,hErr),cCE);
hold off
ylim([-60 40])
xlabel('Temporal Resolution [s]')
ylabel('Percent Error')
title('ve')
subplot(1,4,4)
hold on
shadedErrorPlot(TRes(xRange)',qVpET(xRange,mySig,3),qVpET(xRange,mySig,lErr),qVpET(xRange,mySig,hErr),cET);
shadedErrorPlot(TRes(xRange)',qVpCE(xRange,mySig,3),qVpCE(xRange,mySig,lErr),qVpCE(xRange,mySig,hErr),cCE);
hold off
ylim([-100 300])
xlabel('Temporal Resolution [s]')
ylabel('Percent Error')
title('vp')
%% Make the Error Maps
% These steps should've been moved to a new script, but were appended here
% The next steps load a specific dataset (noiseless, TemporalRes = 1s)
% and makes the percent error maps for kep (Fig 1 in manuscript)
% and computes the error for RRM at vp=0.01 (mentioned in manuscript text)
% Pick which dataset to load
indTRes = 1; % Default (1) : Temporal resolutions = 1 s
indSigma = 1; % Default (1) : Noiseless
repF = 100; % Default (100) | Replications for each noise
curFile = ['Downsample-Noise-' num2str(indSigma) '-TRes-' num2str(indTRes) '.mat'];
load(fullfile(outDir, 'ETM', curFile));
load(fullfile(outDir, 'CERRM', curFile));
% Apply vpCutOff - unnecessary
pkCERRM(~vpMask,:)=[];
pkERRM(~vpMask,:)=[];
pkLRRM(~vpMask,:)=[];
pkERTM(~vpMask,:)=[];
pkETM(~vpMask,:)=[];
pkRTM(~vpMask,:)=[];
pkTM(~vpMask,:)=[];
ktError = PercentError(ktRR*pkCERRM(:,1),trueKt(:));
veError = PercentError(veRR*pkCERRM(:,2),trueVe(:));
kepError = PercentError(pkCERRM(:,3),trueKep(:));
vpError = PercentError(ktRR*pkCERRM(:,4),trueVp(:));
ktErrorCE = reshape(ktError,[nX nY repF]);
kepErrorCE = reshape(kepError,[nX nY repF]);
veErrorCE = reshape(veError,[nX nY repF]);
vpErrorCE = reshape(vpError,[nX nY repF]);
pkERRM = real(pkERRM);
ktError = PercentError(ktRR*pkERRM(:,1),trueKt(:));
veError = PercentError(veRR*pkERRM(:,2),trueVe(:));
kepError = PercentError(pkERRM(:,3),trueKep(:));
vpError = PercentError(ktRR*pkERRM(:,4),trueVp(:));
ktErrorE = reshape(ktError,[nX nY repF]);
kepErrorE = reshape(kepError,[nX nY repF]);
veErrorE = reshape(veError,[nX nY repF]);
vpErrorE = reshape(vpError,[nX nY repF]);
ktError = PercentError(ktRR*pkLRRM(:,1),trueKt(:));
veError = PercentError(veRR*pkLRRM(:,2),trueVe(:));
kepError = PercentError(pkLRRM(:,3),trueKep(:));
ktErrorR = reshape(ktError,[nX nY repF]);
kepErrorR = reshape(kepError,[nX nY repF]);
veErrorR = reshape(veError,[nX nY repF]);
% Missing axes labels
figure('Position',[300 300 1200 500])
subplot(1,3,1)
showErrMap(logModulus(flipud(mean(kepErrorR,3))),3)
title('RRM - kep Percent Error')
subplot(1,3,2)
showErrMap(flipud(mean(kepErrorE,3)),0.1)
title('ERRM - kep Percent Error')
subplot(1,3,3)
showErrMap(flipud(mean(kepErrorCE,3)),0.1)
title('CERRM - kep Percent Error')
%% Show the median percent error and interquartile for RRM at chosen vp
% Recompute errors from previous loaded data
ktError = PercentError(ktRR*pkLRRM(:,1),trueKt(:));
veError = PercentError(veRR*pkLRRM(:,2),trueVe(:));
kepError = PercentError(pkLRRM(:,3),trueKep(:));
chosenVp = 0.01;
estsKt = ktError(trueVp(:)==chosenVp,:);
estsKep = kepError(trueVp(:)==chosenVp,:);
estsVe = veError(trueVp(:)==chosenVp,:);
disp(['Median percent error and interquartile range for RRM when vp = ' num2str(chosenVp)])
disp('For Ktrans:')
disp([median(estsKt(:)) iqr(estsKt(:))])
disp('For kep:')
disp([median(estsKep(:)) iqr(estsKep(:))])
disp('For ve:')
disp([median(estsVe(:)) iqr(estsVe(:))])