-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathMT_calibration_experiment_figures.m
164 lines (154 loc) · 8.02 KB
/
MT_calibration_experiment_figures.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
%% group analysis
%%% makes figures and tables for MTsat calibration analysis
close all
brain_idx = find(sum(all.MTsat_reference) ~= 0);
colors_to_use = parula(length(brain_names));
%% histograms
for_histogram = {'n','R2','A','C','MTsat_reference'};
for hist_to_make = 1:length(for_histogram)
f = figure(hist_to_make);
set(f,'Position',[100 100 1000 300]);
varname = for_histogram{hist_to_make}
clear lower_lim upper_lim fi xi
for brain = 1:length(brain_names) %brain_idx
try
values = all.(varname)(:,brain);
idx = find(all.voxelmask(:,brain) ~= 0); %%%
values_to_use = values(idx);
%lower_lim(brain) = prctile(values_to_use,2); %%% limit range of x-axis, otherwise difficult to compare distributions
%upper_lim(brain) = prctile(values_to_use,98);
lower_lim(brain) = 0; %%% limit range of x-axis, otherwise difficult to compare distributions
upper_lim(brain) = max(values_to_use);
mask = find(values_to_use >= lower_lim(brain) & values_to_use <= upper_lim(brain));
test = values_to_use(mask);
[fi, xi] = ksdensity(test); %%% make Gaussians, so values can occur outside the range
plot(xi,fi,'color',colors_to_use(brain,:),'LineWidth',2);
hold on
set(gca,'FontSize',15)
end
end
xlim([min(lower_lim), max(upper_lim)]);
if strcmp(varname,'C')
aah = xline(1.2);
aah.Annotation.LegendInformation.IconDisplayStyle = 'off';
xticks([0:0.2:2])
end
legend(brain_names(brain_idx),'interpreter','None','Location','NorthWest');
xlabel(varname,'interpreter','None');
ylabel('ProbabilityDensity')
saveas(f,[outpath,'/Figure_',variation,'_',varname,'_histo.png'])
end
%% table
curr_c = 'C';
display('value table')
table_filename = [outpath,'/Table_values_',curr_c,'_',variation,'.txt'];
delete(table_filename)
fid = fopen(table_filename,'w');
%% get values for each brain
clear med mean_R2 std_R2
for brain = brain_idx
values = all.(curr_c)(:,brain);
%idx = find(all.ft_reference(:,brain) > 0 ); %%% define brain mask
idx = find(all.voxelmask(:,brain) > 0 ); %%% define brain mask
idx = find(all.voxelmask(:,brain) > 0 & all.C(:,brain) > 0 & all.C(:,brain) < 1.4 ); %%% constrain statistics to realistic values!
%med(brain) = median(values(idx)); %%% median C across brain
med(brain) = mean(values(idx));
var(brain) = std(values(idx));
var_perc(brain) = 100 * var(brain) / med(brain);
pc = partialcorr([all.(curr_c)(idx,brain),all.ft_reference(idx,brain),all.MTsat_reference_const_corr_indopt(idx,brain)],'Type','Spearman','rows','complete');
pc_CvsB1 = pc(1,2);
pc_CvsMT = pc(1,3);
%%% fit median
%mean_R2(brain) = median(all.R2(idx,brain));
%std_R2(brain) = iqr(all.R2(idx,brain));
mean_R2(brain) = mean(all.R2(idx,brain));
std_R2(brain) = std(all.R2(idx,brain));
within_vox_mean(brain) = mean(all.C_SE(idx,brain));
within_vox_std(brain) = std(all.C_SE(idx,brain));
within_vox_mean_perc(brain) = 100 * nanmean(all.C_SE(idx,brain)./ all.C(idx,brain));
within_vox_std_perc(brain) = 100 * nanstd(all.C_SE(idx,brain) ./ all.C(idx,brain));
fprintf(fid,['%s & ',repmat('%.3f%s%.3f & ',1,5),'%.3f & %.3f','%s'],brain_names{brain},med(brain),'$\pm$',var(brain),100,'$\pm$',var_perc(brain),within_vox_mean(brain),'$\pm$',within_vox_std(brain), within_vox_mean_perc(brain),'$\pm$', within_vox_std_perc(brain),mean_R2(brain),'$\pm$',std_R2(brain),pc_CvsB1,pc_CvsMT,'\\');
fprintf(fid,'\n');
end
table_filename = [outpath,'/Table_group_median_iqr_',curr_c,'_',variation,'.txt'];
delete(table_filename)
fid = fopen(table_filename,'w');
fprintf(fid,['%.3f %.3f'], median(med(med~=0)), iqr(med(med~=0)));
fprintf(fid,'\n');
fclose(fid)
table_filename = [outpath,'/Table_group_mean_std_',curr_c,'_',variation,'.txt'];
delete(table_filename)
fid = fopen(table_filename,'w');
fprintf(fid,['%.3f %.3f'], mean(med(med~=0)), std(med(med~=0)));
fprintf(fid,'\n');
fclose(fid)
%% table for comparing corrected maps
display('correction approach comparison table')
correction_approaches = {'uncorrected', 'voxel', 'individual', 'global'};
correction_name_in_structure = {'MTsat_reference','MTsat_reference_corr','MTsat_reference_const_corr_indopt','MTsat_reference_const_corr_groupopt'};
table_filename = [outpath,'/Table_correction_comparison_',curr_c,'_',variation,'.txt'];
delete(table_filename)
fid = fopen(table_filename,'w');
fprintf(fid,['%s'],'brain & uncorr & voxel & ind & group & uncorr vs voxel & uncorr vs ind & voxel vs ind & ind vs group');
fprintf(fid,'\n');
for brain = brain_idx
%figure()
idx = find(all.ft_reference(:,brain) > 0 ); %%% define brain mask
fprintf(fid,['%s & '],brain_names{brain});
%% calculate partial spearman correlation between mt map and b1, regressing out r2s
for c = 1:length(correction_approaches)
correction_approach = correction_approaches{c};
curr_corrected_map_vec = all.(correction_name_in_structure{c})(:,brain); %%% get current map
b1partcorr.(correction_approach)(brain) = partialcorr(curr_corrected_map_vec(idx),all.ft_reference(idx, brain), all.R2s(idx, brain),'type','Spearman');
fprintf(fid,['%.3f & '], b1partcorr.(correction_approach)(brain));
end
%% calculate difference between the different maps
combinations = {[1 2], [1 3], [2 3], [3 4]};
for c2 = 1:length(combinations)
vec1 = all.(correction_name_in_structure{combinations{c2}(1)})(idx,brain); %%% get first map
vec2 = all.(correction_name_in_structure{combinations{c2}(2)})(idx,brain); %%% get second map
percent_change = 100 * abs((vec1 - vec2) ./ vec1);
comparison_name{c2} = [correction_approaches{combinations{c2}(1)},'_vs_',correction_approaches{combinations{c2}(2)}];
absolute_diff_between_maps.(comparison_name{c2})(brain) = nanmedian(percent_change); %nanmean(percent_change);
absolute_diff_between_maps_iqr.(comparison_name{c2})(brain) = iqr(percent_change); %nanstd(percent_change);
absolute_diff_between_maps_max.(comparison_name{c2})(brain) = max(percent_change);
fprintf(fid,['%.2f%s%.2f%s%.2f%s & '], absolute_diff_between_maps.(comparison_name{c2})(brain), '$\pm$', absolute_diff_between_maps_iqr.(comparison_name{c2})(brain), ' \newline (max = ', round(absolute_diff_between_maps_max.(comparison_name{c2})(brain),2),')');
end
fprintf(fid,'\n');
end
fclose(fid)
%% correlation with MT and b1
corrvars = {'C', 'ft_reference', 'MTsat_reference_const_corr_indopt'}
for var1 = 1:length(corrvars)
var1name = corrvars{var1};
lower_lim1 = min(prctile(all.(var1name)(:,brain),5));
upper_lim1 = min(prctile(all.(var1name)(:,brain),95));
for var2 = 1:length(corrvars)
f5 = figure()
set(f5,'position',[1 1 1500 1000]); %, 'units', 'normalized')
count = 1;
for brain = brain_idx
subplot(2,3,count); count = count + 1;
var2name = corrvars{var2};
lower_lim2 = min(prctile(all.(var2name)(:,brain),2));
upper_lim2 = min(prctile(all.(var2name)(:,brain),98));
idx = find(all.(var1name)(:,brain) > lower_lim1 & all.(var1name)(:,brain) < upper_lim1 & all.(var2name)(:,brain) > lower_lim2 & all.(var2name)(:,brain) < upper_lim2);
binscatter(all.(var1name)(idx, brain),all.(var2name)(idx,brain))
xlabel(var1name)
ylabel(var2name)
end
set(f5,'position',[1 1 1000 800]); %, 'units', 'normalized')
saveas(f5,[outpath,'/Figure_',variation,'_',var1name,'_v2_',var2name,'_scatters.png']);
end
end
%% temperature table
display('temperature')
table_filename = [outpath,'/Table_temperature_',variation,'.txt'];
delete(table_filename)
fid = fopen(table_filename,'w');
%% get values for each person
for brain = brain_idx
fprintf(fid,['%s & ',repmat('%.0f & ',1,2),' %s'],brain_names{brain},all.min_temp(brain),all.max_temp(brain),'\\');
fprintf(fid,'\n');
end
fclose(fid)