/
beta_extraction_voi_10.m
217 lines (178 loc) · 6.59 KB
/
beta_extraction_voi_10.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
%% Extract beta estimates for 10 Intensity levels
close all
clear all
clc
%% Data info
SJs = { 'S01' 'S02' 'S03' 'S04' 'S05' 'S06' 'S08' 'S09' 'S11' 'S13' 'S14' 'S15' 'S16' 'S17' 'S18' 'S19' 'S20' 'S21' 'S22' 'S23' 'S24' 'S25' 'S26' 'S28' 'S29' 'S30' 'S32'};
data_dir = 'my_data_dir';
model_dir = 'BayesGLM';
trg_dir = fullfile(data_dir, '10Int betas');
if ~exist(trg_dir,'dir')
mkdir(trg_dir)
end
name = 'Betas_4mmBMSPeakVOIs_10Int';
info = 'Beta estimates for 10 intensity levels from 4mm VOIs centered on individual BMS peaks within group level BMS ROIs, mean estimates';
% Bayesian 1st level models that define VOIs
models = {'Int'
'PF'
'Det'
'Unc'
'Rep'};
% Bayesian 1st level model to extract parameters from
ffx = 'BayesFFXmpctd_10RT';
% VOI name model
VOIs = { 'BMS_Int_rSI_4mm' 'R SIa' 1
'BMS_Int_rSIIa_4mm' 'R SIIa' 1
'BMS_Int_rSIIp_4mm' 'R SIIp' 1
'BMS_Int_lSII_4mm' 'L SIIm' 1
'BMS_PF_rSI_4mm' 'R SIp' 2
'BMS_PF_rSII_4mm' 'R SII' 2
'BMS_PF_lSII_4mm' 'L SII' 2
'BMS_Det_rSIIs_4mm' 'R SIIs' 3
'BMS_Det_rSIIi_4mm' 'R SIIi' 3
'BMS_Det_lSII_4mm' 'L SIIl' 3
'BMS_Det_lIPS_4mm' 'L IPL' 3
'BMS_Det_lMFG_4mm' 'L SFG' 3
'BMS_Det_lLG_4mm' 'L V3' 3
'BMS_UncMSF_4mm' 'SMG/ACC' 4
'BMS_UncrAIC_4mm' 'R AIC' 4
'BMS_UnclAIC_4mm' 'L AIC' 4
'BMS_ReplSMA_4mm' 'L SMA' 5
'BMS_ReplThal_4mm' 'L Thal' 5
'BMS_ReprSMG_4mm' 'R SMaG' 5 };
nVOI = size(VOIs,1);
nSubs = numel(SJs);
nInt = 10;
nRuns = 4;
voi_dirs = cellfun(@strcat,repmat({'VOIs\FFXmpctd_'},nVOI,1),models([VOIs{:,3}]),repmat({'RT'},nVOI,1),'UniformOutput',0);
% beta images
prefix_Cbeta = 'rCbeta_';
b = 1:4:37; % index beta images of regressors of interest
%% Get estimates
P = cell2struct(cell(size(VOIs,1),1),VOIs(:,1),1);
for v = 1:nVOI
voi = VOIs{v,1};
disp(voi)
if exist(fullfile(trg_dir,[name '_' voi '.mat']),'file')
load(fullfile(trg_dir,[name '_' voi '.mat']))
fns = fieldnames(betas);
val = struct2cell(betas);
P.(VOIs{v,1}) = cell2struct(val,fns);
disp(['Loaded ' name 'betas into P struct!'])
else
% Prepare data structure
P.(VOIs{v,1}).pM = nan(nInt, nRuns, nSubs);
P.(VOIs{v,1}).info = info;
P.(VOIs{v,1}).voi = VOIs{v,1};
% Assemble mean estimates for all voxels in VOIs, subjects, runs, and intensities
for s = 1:nSubs
disp(SJs{s})
% get betas
sub_dir = fullfile(data_dir, SJs{s}, model_dir, ffx);
voi_file = fullfile(data_dir, SJs{s}, voi_dirs{v}, ['VOI_' SJs{s} '_' voi '_mask.nii']);
for r = 1:nRuns
disp(r)
for i = 1:nInt
beta_idx = b(i)+(r-1)*56;
Cbeta_img = spm_select('FPList', sub_dir, ['^' prefix_Cbeta '0{1,3}' num2str(beta_idx) '.nii$']);
P.(VOIs{v,1}).pM(i,r,s) = spm_summarise(Cbeta_img, voi_file, @mean);
end
end
end
% Aggregate runs
P.(voi).pM_r = squeeze(mean(P.(voi).pM,2))';
% Aggregate subjects
P.(voi).pM_rs = mean(P.(voi).pM_r,1);
P.(voi).SD_rs = std(P.(voi).pM_r,1);
P.(voi).SE_rs = P.(voi).SD_rs/sqrt(nSubs);
end
end
%% Plot
% Regressors
% intensity
INT = (1:10)';
% detection
det = fullfile('mydir','normDet.mat');
load(det)
DET = mean(det)';
% pfs
pfs = fullfile('mydir','normPFs.mat');
load(pfs)
PF = mean(pfs)';
% uncertainty
unc = fullfile('mydir','normUnc.mat');
load(unc)
UNC = mean(unc)';
% report
REP = ones(10,1);
reg = { INT
PF
DET
UNC
REP };
cols = {[0 1 0] % Green: Intensity
[0 0 1] % Blue: P(Detection)
[1 0 0] % Red: Detection
[0 1 1] % Cyan: Uncertainty
[1 0 1]}; % Magenta: Report
for v = 1:nVOI
voi = VOIs{v,1};
% Get data
beta = P.(voi).pM_rs;
SEbeta = P.(voi).SE_rs;
% Get model
[b,dev,stats] = glmfit(reg{VOIs{v,3}},beta','normal');
mod = b(2)*reg{VOIs{v,3}}+b(1);
f = figure;
hold on
set(gca,'FontSize',10,'FontName','Calibri')
% Plot data and model
plot(mod,'LineWidth',2,'Color', cols{VOIs{v,3}})
ha = errorbar(beta,SEbeta,'ko','MarkerSize',3,'MarkerFaceColor',[0 0 0]);
% Remove horizonzal lines from error bars
hb = get(ha,'children');
Xdata = get(hb(2),'Xdata');
temp = 4:3:length(Xdata);
temp(3:3:end) = [];
xleft = temp; xright = temp+1;
Xdata(xleft) = 0;
Xdata(xright) = 0;
set(hb(2),'Xdata',Xdata)
% Axes and Ticks
top = beta+SEbeta;
bottom = beta-SEbeta;
min_b = min(bottom);
max_b = max(top);
range_b = max_b - min_b;
margin = range_b/20;
ylim([(min_b - margin) (max_b + margin)])
ylim_min = (ceil(min_b*10)/10);
ylim_max = (floor(max_b*10)/10);
clear mod
if mod(round((ylim_max-ylim_min)*10),2) ~= 0
ylim_min = ylim_min+.1;
end
ylabs = [ylim_min (ylim_max - (ylim_max-ylim_min)/2) ylim_max];
set(gca,'YTick',ylabs)
ylabs_str = cellstr(num2str(ylabs'));
ylabs_str = cellfun(@strrep,ylabs_str, repmat({'0.'},numel(ylabs_str),1), repmat({'.'},numel(ylabs_str),1),'UniformOutput',0);
set(gca,'YTickLabel',{})
xlims = xlim;
text(repmat(xlims(1)-0.01,numel(ylabs_str),1),ylabs,ylabs_str,'HorizontalAlignment','right','FontName','Calibri','FontSize',8)
set(gca,'XTick',[])
set(gca,'xcolor',[1 1 1])
set(gca,'TickLength',[0.02,0.5])
title(VOIs{v,2})
box off
% Figure size
x0 = 1;
y0 = 1;
width = 1.5;
height = 2;
set(gcf,'Units','centimeters','position',[0,0,14,21])
set(gca,'units','centimeters','position',[x0,y0,width,height])
% Save
saveas(f,fullfile(trg_dir,[name,'_',VOIs{v,1},'.fig']))
betas = P.(voi);
save(fullfile(trg_dir,[name,'_',VOIs{v,1},'.mat']),'betas')
end