/
C6_Plotting_classification_and_feature_importance.m
549 lines (527 loc) · 19.7 KB
/
C6_Plotting_classification_and_feature_importance.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
% This code plots the AUCs obtained from the classifications
% of contacts in 6 classifications including:
% Within-patient classification (interictal and ictal)
% Across-patient generalisation (interictal and ictal)
% Across-time generalisation (interictal to ictal and vice versa)
% INPUTS: data from C5_Permuting_all_classifciation_windows
% OUTPUTS: figures and numbers produced in command window
%%
clc;
clear all;
close all;
addpath(genpath('bayesFactor-master'))
load('random_permutations.mat')
what=2;% 1=within patient; 2= across time; 3=across patients
Measures={'accuracy','sensitivity','specificity','precision','recall','f1-measure','gmean','auc'};
metric=8; %AUC
if what==1
data_inter=Performance_within_inter(:,metric);
data_ictal=Performance_within_ictal(:,metric);
datas_random_inter=squeeze(nanmean(Performance_rand_within_inter(:,metric,:),3));
datas_random_ictal=squeeze(nanmean(Performance_rand_within_ictal(:,metric,:),3));
elseif what==2
data_inter=Performance_acrss_time_inter(:,metric);
data_ictal=Performance_acrss_time_ictal(:,metric);
datas_random_inter=squeeze(nanmean(Performance_rand_acrss_time_inter(:,metric,:),3));
datas_random_ictal=squeeze(nanmean(Performance_rand_acrss_time_ictal(:,metric,:),3));
elseif what==3
data_inter=Performance_acrss_subj_inter(:,metric);
data_ictal=Performance_acrss_subj_ictal(:,metric);
datas_random_inter=squeeze(nanmean(Performance_rand_acrss_subj_inter(:,metric,:),3));
datas_random_ictal=squeeze(nanmean(Performance_rand_acrss_subj_ictal(:,metric,:),3));
end
datas=[data_inter data_ictal];
datas_random=[datas_random_inter datas_random_ictal];
colours={[0 0 0],[0.3 0.3 0.3]};
datas_tmp_outed=nan(55,2);
for i=1:2
tmpp=rmoutliers(datas(:,i));
datas_tmp_outed(1:length(tmpp),i)=tmpp;
swarmchart([i*ones(length(tmpp),1)],tmpp,'MarkerFaceColor',colours{i},'MarkerEdgeColor',colours{i},'MarkerFaceAlpha',0.3,'MarkerEdgeAlpha',0.3);
hold on;
end
boxplot(datas_tmp_outed,'Whisker',inf,'color','k');
xticks([1 2])
ylabel('Area Under Curve (AUC)')
box off
grid on
set(gca,'TickDir','out','Fontsize',16)
if what==1
title('Within-Patient Classification')
xticklabels({'Interictal';'Ictal'})
ylim([0.35 1])
% ylim([0.8 1])
% yticks([0.8 0.9 1])
% yticklabels({'0.8','0.95','1'})
elseif what==2
title('Cross-Time Generalisation')
xticklabels({'Interictal-to-Ictal';'Ictal-to-Interictal'})
ylim([0.35 1])
yticks([0.4:0.1:1])
elseif what==3
title('Cross-Patient Generalisation')
xticklabels({'Interictal';'Ictal'})
ylim([0.35 1])
yticks([0.4:0.1:1])
end
plot([0.5 2.5],[0.5 0.5],'--k')
bays_against_chance=[bf.ttest2(datas(:,1),datas_random(:,1)) bf.ttest2(datas(:,2),datas_random(:,2))]
bays_diff=bf.ttest(datas(:,1),datas(:,2))
[mean(datas(:,1)) std(datas(:,1)) mean(datas(:,2)) std(datas(:,2))]
%% Features importance
clc
close all
feats=[1:8 10:12 14:36];
Features_labels={'Mean','Median','Variance','Skewness','Kurtosis',...
'LZ Comp','Higichi FD','Katz FD','Lyap Exp','Hurst Exp',...
'Samp Ent','Apprx Ent','Autocorr','Hjorth Comp','Hjorth Mob',...
'Mean Freq','Med Freq','Avg Freq','SEF','Pow Med Freq',...
'Phs Med Freq','Power','Energy Ratio','Delta Pow','Theta Pow',...
'Alpha Pow','Beta Pow','Gamma Pow','H-Gamma Pow','Correlation','Delta Coh',...
'Theta Coh','Alpha Coh','Beta Coh','Gamma Coh','H-Gamma Coh'};
what=2;% 1=within patient; 2= across time; 3=across patients
ictal_or_inter='Interictal';
sorted=0;
if what==1
load(['Within_subject_performance_sepnorm',ictal_or_inter,'_all_feats_crcted_feats_imp_100_10.mat'])
elseif what==2
if strcmp(ictal_or_inter,'Interictal')
load(['Generalisation_performance_across_time_all_feats_comb_crcted_feats_ovrsmp_imp_100_10.mat'])
else
load(['Generalisation_performance_across_time_ict_to_int_all_feats_comb_crcted_feats_ovrsmp_imp_100_10.mat'])
end
elseif what==3
load(['Generalisation_performance_across_subjects_',ictal_or_inter,'_all_feats_comb_crcted_feats_ovrsmp_imp_100_10.mat'])
end
ff=1;
iter_rand=1;
feat_imp=nan(1,34,1);
for fld=1:size(impCART,5)
for iter_equalis=1
for p=1:55
feat_imp(p,included_feats{ff,iter_equalis,p,iter_rand},iter_equalis,fld)=impCART{ff,iter_equalis,p,iter_rand,fld};
end
end
end
feat_imp(feat_imp==0)=nan;
figure;
datas=nanmean(feat_imp,4);
datas_tmp_outed=nan(55,34);
orders=1:size(datas,2);
if sorted==1
for i=1:size(datas,2)
tmpp=rmoutliers(datas(:,i));
datas_tmp_outed(1:length(tmpp),i)=tmpp;
end
[~,order]=sort(nanmedian(datas_tmp_outed),'descend');
% [~,order]=sort(nanmean(datas_tmp_outed),'descend');
orders=order;
end
c=0;
datas_tmp_outed=nan(55,34);
for i=orders
c=c+1;
if i<6
colours=[0.7 0.7 0.5];
elseif i>5 && i<14
colours=[0.9 0.5 0.9];
elseif i>13 && i<28
colours=[0.3 0.7 0.7];
else
colours=[0.5 0.5 0.9];
end
tmpp=rmoutliers(datas(:,i));
datas_tmp_outed(1:length(tmpp),c)=tmpp;
swarmchart([c*ones(size(tmpp,1),1)]',tmpp,20,'MarkerFaceColor',colours,'MarkerEdgeColor',colours,'MarkerFaceAlpha',0.5,'MarkerEdgeAlpha',0.3);
hold on;
clearvars tmpp
end
boxplot(datas_tmp_outed,'Whisker',inf,'color','k');
box off
grid on
set(gca,'TickDir','out','Fontsize',16)
ylabel('Feature Contribution (A.U.)')
xticks([1:length(feats)])
if sorted==1
xticklabels([Features_labels(feats(order))])
else
xticklabels([Features_labels(feats)])
end
if what==1
ylim([-0.5 2.5])
title(['Within-Patient Classification ',ictal_or_inter])
elseif what==2
ylim([-0.5 4])
if strcmp(ictal_or_inter,'Interictal')
title(['Cross-Time Generalisation Interictal to Ictal',])
else
title(['Cross-Time Generalisation Ictal to Interictal',])
end
elseif what==3
ylim([1 8])
title(['Cross-Patient Generalisation ',ictal_or_inter])
end
%% BF Anova 4 factors
clc;
clear all;
close all;
addpath(genpath('bayesFactor-master'))
load('random_permutations.mat')
participants_info=tdfread(['F:\RESEARCH\Hamid\Multicentre dataset\ds004100\participants.tsv']);
subjects_analysed=[1:25 27:40 42:45 47:58]; % already remove 41 becuase "data" does not exist
what=3;% 1=within patient; 2= across time; 3=across patients
ictal_or_inter='Interictal'; % 1=ictal; non-1=inter
Measures={'accuracy','sensitivity','specificity','precision','recall','f1-measure','gmean','auc'};
metric=8; %AUC
if what==1
if strcmp(ictal_or_inter,'Interictal')
data_tmp=Performance_within_inter(:,metric);
data_tmp_random=squeeze(nanmean(Performance_rand_within_inter(:,metric,:),3));
else
data_tmp=Performance_within_ictal(:,metric);
data_tmp_random=squeeze(nanmean(Performance_rand_within_ictal(:,metric,:),3));
end
elseif what==2
if strcmp(ictal_or_inter,'Interictal')
data_tmp=Performance_acrss_time_inter(:,metric);
data_tmp_random=squeeze(nanmean(Performance_rand_acrss_time_inter(:,metric,:),3));
else
data_tmp=Performance_acrss_time_ictal(:,metric);
data_tmp_random=squeeze(nanmean(Performance_rand_acrss_time_ictal(:,metric,:),3));
end
elseif what==3
if strcmp(ictal_or_inter,'Interictal')
data_tmp=Performance_acrss_subj_inter(:,metric);
data_tmp_random=squeeze(nanmean(Performance_rand_acrss_subj_inter(:,metric,:),3));
else
data_tmp=Performance_acrss_subj_ictal(:,metric);
data_tmp_random=squeeze(nanmean(Performance_rand_acrss_subj_ictal(:,metric,:),3));
end
end
pt=0;
for P=subjects_analysed
pt=pt+1;
% g1{pt,1}=participants_info.engel(P,1);
g1{pt,1}=participants_info.outcome(P,1);
g2{pt,1}=participants_info.target(P,:);
g3{pt,1}=participants_info.lesion_status(P,:);
g4{pt,1}=participants_info.implant(P,:);
end
clearvars data_ready
nans=isnan(data_tmp);
nans([1])=1; % bad labeling of lesion (1)
nans([4 25 30 37 38])=1; % resection targets with fewer than 5 patients
data_ready=table(data_tmp,g1,g2,g3,g4);
data_ready(nans,:)=[];
data_ready.Properties.VariableNames = {'Effect' 'Outcome' 'ROI' 'Lesion' 'Recording'};
bfFull= bf.anova(data_ready,'Effect ~ Outcome+ROI+Lesion+Recording');
bfRestricted= bf.anova(data_ready,'Effect ~ ROI+Lesion+Recording');
effects(1) = bfFull/bfRestricted;
bfRestricted= bf.anova(data_ready,'Effect ~ Outcome+Lesion+Recording');
effects(2) = bfFull/bfRestricted;
bfRestricted= bf.anova(data_ready,'Effect ~ Outcome+ROI+Recording');
effects(3) = bfFull/bfRestricted;
bfRestricted= bf.anova(data_ready,'Effect ~ Lesion+Outcome+ROI');
effects(4) = bfFull/bfRestricted;
% Plotting accuracies across variables
for condition=1:4
clearvars data data_random
figure
feats=1;
data(feats,:)=data_tmp;
data_random(feats,:)=data_tmp_random;
cond=nan(size(data,2),1);
if condition==1
% % strings={'1','2','3','4'};
ss={'S','F'};
strings={'Engel I','Engel II-IV'};
col=cool;
shiftee=0.1;
s=0;
for i=subjects_analysed
s=s+1;
for c=1:length(ss)
% if contains(participants_info.engel(s,1),strings{c})
if contains(participants_info.outcome(s,1),ss{c})
cond(s)=(c);
end
end
end
elseif condition==2
% strings={'FRONTAL','TEMPORAL','FP','MTL','PARIETAL','MFL','INSULAR'};
strings={'FRONTAL','TEMPORAL','MTL'};
col=jet;
shiftee=0.2;
s=0;
for i=subjects_analysed
s=s+1;
for c=1:length(strings)
if contains(participants_info.target(s,:),strings{c})
cond(s)=(c);
end
end
end
cond(isnan(cond))=[];
elseif condition==3
strings={'LESIONAL','NON-LESIONAL'};
col=cool;
shiftee=0.1;
s=0;
for i=subjects_analysed
s=s+1;
for c=1:length(strings)
if contains(participants_info.lesion_status(s,:),strings{c})
cond(s)=(c);
end
end
end
cond(1)=[]; % unused
data(1)=[];
data_random(1)=[];
elseif condition==4
strings={'SEEG','ECOG'};
col=cool;
shiftee=0.1;
s=0;
for i=subjects_analysed
s=s+1;
for c=1:length(strings)
if contains(participants_info.implant(s,:),strings{c})
cond(s)=(c);
end
end
end
end
feats=1;
bar_width=0.2;
tmp=round(linspace(1,length(col),length(strings)));
cols={col(tmp,:)};
multip=floor(length(unique(cond))/2)*0.2;
uniques=unique(cond);
datas=nan(55,length(uniques));
c=0;
datas=nan(55,length(uniques));
for i=uniques'
c=c+1;
cond(40)=0; % unused
data_tmp_tmp=rmoutliers(data(feats,cond==i));
data_tmp_tmp_random=data_random(feats,cond==i);
datas(1:length(data_tmp_tmp),c)=data_tmp_tmp;
if strcmp(ictal_or_inter,'Interictal')
colours=[0 0 0];
else
colours=[0.3 0.3 0.3];
end
swarmchart([c*ones(length(data_tmp_tmp),1)]',data_tmp_tmp,'MarkerFaceColor',colours,'MarkerEdgeColor',colours,'MarkerFaceAlpha',0.5,'MarkerEdgeAlpha',0.3);
hold on;
bays_against_chance{condition,c}=bf.ttest2(data_tmp_tmp,data_tmp_tmp_random);
end
[nanmean(datas) nanstd(datas)]
boxplot(datas,'Whisker',inf,'color','k');
box off
grid on
set(gca,'TickDir','out','Fontsize',16)
xticks([1:size(datas,2)])
xlim([0.5 size(datas,2)+0.5])
plot([0.5 size(datas,2)+0.5],[0.5 0.5],'--k')
title(['ANOVA BF = ',num2str(effects(condition))])
if what==1
ylabel('Within-Patient Classification (AUC)')
% ylim([0.8 1.0])
% yticks([0.8 0.9 1])
ylim([0.35 1.0])
yticks([0.4:0.1:1])
elseif what==2
ylabel('Cross-Time Generalisation (AUC)')
ylim([0.35 1])
yticks([0.4:0.1:1])
elseif what==3
ylabel('Cross-Patient Generalisation (AUC)')
ylim([0.35 1])
yticks([0.4:0.1:1])
end
if condition==2
% strings={'FRT','TPR','FPL','MTL','PRT','MFL','INS'};
strings={'FRT','TPR','MTL'};
end
xticklabels(strings)
if condition==1
% xlabel('Outcome (Engel Level)')
xlabel('Outcome')
elseif condition==2
xlabel('Region of Resection')
elseif condition==3
xlabel('Pathology')
elseif condition==4
xlabel('Recording')
end
clearvars datas
% BFs across conditions
pt=0;
for P=subjects_analysed
pt=pt+1;
if condition==1
% g{pt,1}=participants_info.engel(P,1);
g{pt,1}=participants_info.outcome(P,1);
elseif condition==2
g{pt,1}=participants_info.target(P,:);
elseif condition==3
g{pt,1}=participants_info.lesion_status(P,:);
elseif condition==4
g{pt,1}=participants_info.implant(P,:);
end
end
data_bf=data;
nans=isnan(data_bf);
nans([1])=1; % bad labeling of lesion (1)
nans([4 25 30 37 38 39])=1; % resection targets with fewer than 5 patients
data_bf(nans)=[];
g(nans)=[];
uniq_conds_tmp=unique(g);
combs_tmp=nchoosek(1:length(uniq_conds_tmp),2);
for comb=1:size(combs_tmp,1)
try
conditions_BFtmp(comb,1)=bf.ttest2(data_bf(strcmp(g,uniq_conds_tmp(combs_tmp(comb,1)))),data_bf(strcmp(g,uniq_conds_tmp(combs_tmp(comb,2)))));
catch
conditions_BFtmp(comb,1)=nan;
end
combinations_tmp{comb,1}=uniq_conds_tmp(combs_tmp(comb,1));
combinations_tmp{comb,2}=uniq_conds_tmp(combs_tmp(comb,2));
end
combinations_BF{condition}=conditions_BFtmp;
combinations{condition}=combinations_tmp;
clearvars data_bf conditions_BFtmp combinations_tmp
end
%% Features importance-Outcome anova
clc
close all
feats=[1:8 10:12 14:36];
Features_labels={'Mean','Median','Variance','Skewness','Kurtosis',...
'LZ Comp','Higichi FD','Katz FD','Lyap Exp','Hurst Exp',...
'Samp Ent','Apprx Ent','Autocorr','Hjorth Comp','Hjorth Mob',...
'Mean Freq','Med Freq','Avg Freq','SEF','Pow Med Freq',...
'Phs Med Freq','Power','Energy Ratio','Delta Pow','Theta Pow',...
'Alpha Pow','Beta Pow','Gamma Pow','H-Gamma Pow','Correlation','Delta Coh',...
'Theta Coh','Alpha Coh','Beta Coh','Gamma Coh','H-Gamma Coh'};
participants_info=tdfread(['F:\RESEARCH\Hamid\Multicentre dataset\ds004100\participants.tsv']);
subjects_analysed=[1:25 27:40 42:45 47:58]; % already remove 41 becuase "data" does not exist
what=3;% 1=within patient; 2= across time; 3=across patients
ictal_or_inter='Interictal';
sorted=0;
if what==1
load(['Within_subject_performance_sepnorm',ictal_or_inter,'_all_feats_crcted_feats_imp_100_10.mat'])
elseif what==2
if strcmp(ictal_or_inter,'Interictal')
load(['Generalisation_performance_across_time_all_feats_comb_crcted_feats_ovrsmp_imp_100_10.mat'])
else
load(['Generalisation_performance_across_time_ict_to_int_all_feats_comb_crcted_feats_ovrsmp_imp_100_10.mat'])
end
elseif what==3
load(['Generalisation_performance_across_subjects_',ictal_or_inter,'_all_feats_comb_crcted_feats_ovrsmp_imp_100_10.mat'])
end
ff=1;
iter_rand=1;
feat_imp=nan(1,34,1);
for fld=1:size(impCART,5)
for iter_equalis=1
for p=1:55
feat_imp(p,included_feats{ff,iter_equalis,p,iter_rand},iter_equalis,fld)=impCART{ff,iter_equalis,p,iter_rand,fld};
end
end
end
feat_imp(feat_imp==0)=nan;
datas=nanmean(feat_imp,4);
data_plot=nan(1,34,55,2);
for condition=[1]
for feat=1:34
% BFs across conditions
pt=0;
for P=subjects_analysed
pt=pt+1;
if condition==1
% if contains(participants_info.engel(P,1:2),'1A') || contains(participants_info.engel(P,1:2),'2A')
% g{pt,1}='S';
% else
% g{pt,1}='F';
% end
% % g{pt,1}=participants_info.engel(P,1);
g{pt,1}=participants_info.outcome(P,1);
elseif condition==2
g{pt,1}=participants_info.target(P,:);
elseif condition==3
g{pt,1}=participants_info.lesion_status(P,:);
elseif condition==4
g{pt,1}=participants_info.implant(P,:);
end
end
data_bf=datas(:,feat);
nans=isnan(data_bf);
nans([1])=1; % bad labeling of lesion (1)
data_bf(nans)=[];
g(nans)=[];
uniq_conds_tmp=unique(g);
combs_tmp=nchoosek(1:length(uniq_conds_tmp),2);
for comb=1:size(combs_tmp,1)
try
conditions_BFtmp(comb,1)=bf.ttest2(data_bf(strcmp(g,uniq_conds_tmp(combs_tmp(comb,1)))),data_bf(strcmp(g,uniq_conds_tmp(combs_tmp(comb,2)))));
dat_tmp=data_bf(strcmp(g,uniq_conds_tmp(combs_tmp(comb,1))));
data_plot(comb,feat,1:length(dat_tmp),1)=dat_tmp;
dat_tmp=data_bf(strcmp(g,uniq_conds_tmp(combs_tmp(comb,2))));
data_plot(comb,feat,1:length(dat_tmp),2)=dat_tmp;
catch
conditions_BFtmp(comb,1)=nan;
end
combinations_tmp{comb,1}=uniq_conds_tmp(combs_tmp(comb,1));
combinations_tmp{comb,2}=uniq_conds_tmp(combs_tmp(comb,2));
end
combinations_BF{condition,feat}=conditions_BFtmp;
combinations{condition,feat}=combinations_tmp;
clearvars g data_bf conditions_BFtmp combinations_tmp
end
figure
c=0;
for feat=1:34
c=c+7;
colours=[0.3 0.7 0.3];
data_tmp2=squeeze(data_plot(1,feat,~isnan(data_plot(1,feat,:,2)),2));
swarmchart([(c-1)*ones(sum(~isnan(data_tmp2)),1)]',data_tmp2,'MarkerFaceColor',colours,'MarkerEdgeColor',colours,'MarkerFaceAlpha',0.5,'MarkerEdgeAlpha',0.3);
hold on;
colours=[0.7 0.3 0.3];
data_tmp1=squeeze(data_plot(1,feat,~isnan(data_plot(1,feat,:,1)),1));
swarmchart([(c+1)*ones(sum(~isnan(data_tmp1)),1)]',data_tmp1,'MarkerFaceColor',colours,'MarkerEdgeColor',colours,'MarkerFaceAlpha',0.5,'MarkerEdgeAlpha',0.3);
BFs=bf.ttest2(data_tmp1,data_tmp2);
if what<3
y_position=-0.45;
elseif what==3
y_position=1.05;
end
t=text(c,y_position,sprintf('%0.2f',BFs));
t.Rotation=90;
t.FontSize=14;
if BFs>3
t.FontWeight='Bold';
end
clearvars data_tmp1 data_tmp2
end
box off
grid on
set(gca,'TickDir','out','Fontsize',16)
ylabel('Feature Contributon (A.U.)')
feats=[1:8 10:12 14:36];
xticks([7:7:7*length(feats)])
xticklabels([Features_labels(feats)])
if what==1
ylim([-0.5 2.5])
title(['Within-Patient Classification ',ictal_or_inter])
elseif what==2
ylim([-0.5 4])
if strcmp(ictal_or_inter,'Interictal')
title(['Cross-Time Generalisation Interictal to Ictal',])
else
title(['Cross-Time Generalisation Ictal to Interictal',])
end
elseif what==3
ylim([1 8])
title(['Cross-Patient Generalisation ',ictal_or_inter])
end
end