-
Notifications
You must be signed in to change notification settings - Fork 2
/
scalpmap.m
278 lines (218 loc) · 12.3 KB
/
scalpmap.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
classdef scalpmap
% holds scalpmap data and handles plotting, polarity normalization...
properties
originalChannelWeight = []; % a 3D array (comps, scalp x, scalp y) that holds the topomaps (ICA mixing weights for channels) before polarity normalization
normalizedChannelWeight = []; % a 3D (comps, scalp x, scalp y) array that holds the topomaps (ICA mixing weights for channels)
gridX, gridY = []; % the 2D grid on which scalp maps are plotted.
normalizationPolarity = []; % a vector (containing 1 and -1) that hold polarities for normalizaing scalp maps.
subjectName % a cell array containing subject names associates with dipoles.
numberInDataset % an array containing the number of dipole IC in the associated EEG dataset. for example, if the dipole is from component number 5 in the dataset, this number will be 5. This array is optional.
isNormalized = false % is scalpmaps are currently normalized
end;
properties (Dependent = true)
numberOfScalpmaps
end;
methods
function numberOfScalpmaps = get.numberOfScalpmaps(obj)
numberOfScalpmaps = size(obj.originalChannelWeight,1);
end;
function obj = normalizePolarity(obj, normalizationMethod) % normalize polarities
if nargin < 2
normalizationMethod = 'convex';
end;
obj.normalizationPolarity = normalize_ic_polarity(obj.originalChannelWeight(:,:), normalizationMethod);
obj.normalizedChannelWeight = obj.originalChannelWeight .* repmat(obj.normalizationPolarity', [1, size(obj.originalChannelWeight, 2), size(obj.originalChannelWeight, 3)]);
end;
function newObj = createSubsetForId(obj, subsetId, renormalizeScalpmapPolarity)
if nargin < 3
renormalizeScalpmapPolarity = true;
end;
newObj = obj;
if isempty(obj.originalChannelWeight)
fprintf('Measure Projection Warning: scalp-map is empty but a subset is requested.\n');
else
newObj.originalChannelWeight = obj.originalChannelWeight(subsetId,:,:);
end;
if ~isempty(obj.normalizedChannelWeight)
newObj.normalizedChannelWeight = obj.normalizedChannelWeight(subsetId,:,:);
end;
if ~isempty(obj.normalizationPolarity)
newObj.normalizationPolarity = obj.normalizationPolarity(subsetId);
end;
if ~isempty(newObj.subjectName)
newObj.subjectName = newObj.subjectName(subsetId);
end;
if ~isempty(newObj.subjectName)
newObj.numberInDataset = newObj.numberInDataset(subsetId);
end;
if renormalizeScalpmapPolarity
newObj = newObj.normalizePolarity;
end;
end;
function plot(obj, subsetId, visualizationPolarity, createNewFigure, varargin)
% plot(obj, subsetId, visualizationPolarity, createNewFigure)
if nargin<2 || isempty(subsetId)
subsetId = 1:size(obj.normalizedChannelWeight, 1);
if length(subsetId) > 100
fprintf('There are more than 100 scalp-maps, by default only the first 100 are displayed.\n');
subsetId = 1:100;
end;
elseif islogical(subsetId)
subsetId = find(subsetId); % turn from a logical array to indices.
end;
if nargin<3
visualizationPolarity = 1;
end;
if nargin<4
createNewFigure = true;
end;
numberOfRows = ceil((length(subsetId) * 3/4) .^ 0.5);
numberOfColumns = ceil(numberOfRows * 4/3);
if createNewFigure
figure;
end;
for i=1:length(subsetId)
sbplot(numberOfRows, numberOfColumns, i);
toporeplot(visualizationPolarity * squeeze(obj.normalizedChannelWeight(subsetId(i),:,:)),'plotrad',0.5, 'intrad' , 0.5);
% put information in the axis.
userData.id = subsetId(i);
userData.numberInFigure = i;
set(gca, 'userdata', userData);
set(get(gca, 'children'), 'ButtonDownFcn', @scalpmapCallback, 'hittest', 'on', 'userdata', userData)
set(gca, 'ButtonDownFcn', @scalpmapCallback, 'hittest', 'on')
try
if any(~cellfun(@isempty, obj.subjectName)) && ~isempty(obj.numberInDataset)
titleText = [obj.subjectName{subsetId(i)} ' / IC' num2str(obj.numberInDataset(subsetId(i)))];
title(titleText);
elseif ~isempty(obj.numberInDataset)
titleText = ['IC' num2str(obj.numberInDataset(subsetId(i)))];
title(titleText);
else
title(['ID ' num2str(subsetId(i))]);
end;
catch
end;
end;
colormap jet;
end;
function newObj = horzcat(varargin)
newObj = varargin{1};
for i=2:length(varargin)
newObj.originalChannelWeight = cat(1, newObj.originalChannelWeight, varargin{i}.originalChannelWeight);
% attention: not normnalized yet, woudl need additional normalization
newObj.normalizedChannelWeight = cat(1, newObj.normalizedChannelWeight, varargin{i}.normalizedChannelWeight);
newObj.normalizationPolarity = cat(2, newObj.normalizationPolarity, varargin{i}.normalizationPolarity);
newObj.subjectName = cat(1, newObj.subjectName, varargin{i}.subjectName);
newObj.numberInDataset = cat(1, newObj.numberInDataset, varargin{i}.numberInDataset);
end;
% if new scalpmaps are added, a new normalization is required
if size(newObj.originalChannelWeight,1) > size(varargin{1}.originalChannelWeight,1)
newObj.isNormalized = false;
end;
end;
function [id similarity] = detectEye(obj, threshold, varargin)
% [id similarity] = detectEye(obj, threshold, varargin)
%
% detect eye channels (whose equiv. dipole has fallen inside brain volume)
% by comparing them to a set of hand-picked eye scalpmaps and finding
% the ones that are more similar than a threshold to one of these predefined eye
% scalpmaps.
%
% id conatins the indices of detected eye components.
% similarity contains the similarity of each scalpmap to the
% closest eye component in the eye dataset.
if nargin < 2
threshold = 0.944;
end;
eyeDetector = pr.eyeCatch;
[id similarity] = eyeDetector.detectFromScalpmapObj(obj, threshold);
end;
function plotEye(obj, threshold, varargin)
if nargin < 2
threshold = 0.944;
end;
id = detectEye(obj, threshold);
plot(obj, id);
end;
function obj = addFromChannels(obj, channelWeight, chanlocs, numberInDataset, subjectName, varargin)
% channelWeight is channels x number of scalpmaps
if nargin < 5
subjectName = repmat({''}, size(channelWeight,1), 1);
end;
if nargin < 4
numberInDataset = 1:size(channelWeight,1);
end;
% to remove the extra figure which is created by topoplot()
% we need to first have a list of current figures (to delete ones that are added)
figureHandlesBefore = findobj('type', 'figure');
for i = 1:size(channelWeight, 2)
[hfig grid plotrad Xi Yi] = topoplot(channelWeight(:,i), chanlocs, ...
'verbose', 'off',...
'electrodes', 'on' ,'style','both',...
'plotrad',0.55,'intrad',0.55,...
'noplot', 'on');
obj.gridX = Xi;
obj.gridY = Yi;
gridCube(1,:,:) = grid;
obj.originalChannelWeight = cat(1, obj.originalChannelWeight, gridCube);
obj.subjectName = cat(1, obj.subjectName, subjectName(i));
obj.numberInDataset = cat(1, obj.numberInDataset, numberInDataset(i));
end;
figureHandlesAfter = findobj('type', 'figure');
close(setdiff(figureHandlesAfter, figureHandlesBefore));
obj.normalizedChannelWeight = obj.originalChannelWeight;
% if inputOptions.normalizePolarity
% obj = obj.normalizePolarity(inputOptions.normalizationMethod);
% end;
end;
function [obj sameMap] = removeDuplicates(obj, method, varargin)
if nargin < 2
method = 'hash';
end;
if strcmpi(method, 'hash') % more than 10 times faster duplicate detection with two hashes
channelWeight = obj.originalChannelWeight(:,:);
channelWeight(isnan(channelWeight)) = 0;
sumWeight = sum(channelWeight,2);
sumAbsWeight = sum(abs(channelWeight),2);
% use two hashes and select ones that have the same hashes for both hash types
[c ia ic] = unique(sumAbsWeight);
[c2 ia2 ic2] = unique(sumWeight);
isDuplicateBasedOnHash1 = true(length(sumAbsWeight),1);
isDuplicateBasedOnHash2 = true(length(sumAbsWeight),1);
isDuplicateBasedOnHash1(ia) = false;
isDuplicateBasedOnHash2(ia2) = false;
isDuplicateBasedOnHash1and2 = isDuplicateBasedOnHash1 & isDuplicateBasedOnHash2;
isUnique = ~isDuplicateBasedOnHash1and2;
obj = obj.createSubsetForId(isUnique, false);
else % exact method, much much slower
nanId = isnan(obj.originalChannelWeight(1,:));
channelWeight = obj.originalChannelWeight(:, ~nanId);
sameMap = logical(spalloc(size(channelWeight,1),size(channelWeight,1), size(channelWeight,1)));
progress('init'); % start the text based progress bar
for i=1:size(sameMap, 1)
progress(i / size(sameMap, 1), sprintf('\npercent done %d/100',round(100*i / size(sameMap, 1))));
theSame = channelWeight(i,1) == channelWeight(:,1);
for j=1:size(channelWeight,2)
if any(theSame)
theSame(theSame) = theSame(theSame) & channelWeight(i,j) == channelWeight(theSame,j);
end;
end;
sameMap(i,theSame) = true;
sameMap(theSame,i) = true;
end;
% keep the first of duplicates and removes others
isDuplicate = false(size(sameMap,1),1);
for i=1:size(sameMap,1)
if ~isDuplicate(i)
isDuplicate(sameMap(i,:)) = true;
isDuplicate(i) = false;
end;
end;
obj = obj.createSubsetForId(~isDuplicate, false);
pause(.1);
progress('close'); % duo to some bug need a pause() before
fprintf('\n');
end;
end;
end;
end