/
testGlobalGLCM.m
144 lines (129 loc) · 4.87 KB
/
testGlobalGLCM.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
% this script tests global GLCM texture features between CERR and ITK.
%
% APA, 11/28/2016
% Number of Gray levels
nL = 4;
voxelOffset = 1;
dirFlag = 1;
% Random n x n x n matrix
n = 10;
testM = rand(n,n,n);
testM = imquantize_cerr(testM,nL);
maskBoundingBox3M = testM .^ 0;
% % Structure from planC
% global planC
% indexS = planC{end};
% scanNum = 1;
% structNum = 16;
%
% [rasterSegments, planC, isError] = getRasterSegments(structNum,planC);
% [mask3M, uniqueSlices] = rasterToMask(rasterSegments, scanNum, planC);
% scanArray3M = getScanArray(planC{indexS.scan}(scanNum));
%
% SUVvals3M = mask3M.*double(scanArray3M(:,:,uniqueSlices));
% [minr, maxr, minc, maxc, mins, maxs]= compute_boundingbox(mask3M);
% maskBoundingBox3M = mask3M(minr:maxr,minc:maxc,mins:maxs);
% volToEval = SUVvals3M(minr:maxr,minc:maxc,mins:maxs);
% volToEval(maskBoundingBox3M==0) = NaN;
%
% testM = imquantize_cerr(volToEval,nL);
% % CERR texture
% flagS.energy = 1;
% flagS.entropy = 1;
% flagS.corr = 1;
% flagS.haralickCorr = 1;
% flagS.clustShade = 1;
% flagS.clustProm = 1;
% flagS.contrast = 1;
% flagS.invDiffMoment = 1;
% flagS.sumAvg = 1;
glcmFlagS.energy = 1;
glcmFlagS.jointEntropy = 1;
glcmFlagS.jointMax = 1;
glcmFlagS.jointAvg = 1;
glcmFlagS.jointVar = 1;
glcmFlagS.contrast = 1;
glcmFlagS.invDiffMoment = 1;
glcmFlagS.sumAvg = 1;
glcmFlagS.corr = 1;
glcmFlagS.clustShade = 1;
glcmFlagS.clustProm = 1;
glcmFlagS.haralickCorr = 1;
glcmFlagS.invDiffMomNorm = 1;
glcmFlagS.invDiff = 1;
glcmFlagS.invDiffNorm = 1;
glcmFlagS.invVar = 1;
glcmFlagS.dissimilarity = 1;
glcmFlagS.diffEntropy = 1;
glcmFlagS.diffVar = 1;
glcmFlagS.diffAvg = 1;
glcmFlagS.sumVar = 1;
glcmFlagS.sumEntropy = 1;
glcmFlagS.clustTendency = 1;
glcmFlagS.autoCorr = 1;
glcmFlagS.invDiffMomNorm = 1;
glcmFlagS.firstInfCorr = 1;
glcmFlagS.secondInfCorr = 1;
tic
cooccurType = 2; % generates separate cooccurrance matrix for each direction.
offsetsM = getOffsets(dirFlag) * voxelOffset;
offsetsM = offsetsM(1:2,:);
cooccurM = calcCooccur(testM, offsetsM, nL, cooccurType);% Note: cooccurM is
% ... of size (nLxnL, number of directions)
featureS = cooccurToScalarFeatures(cooccurM, glcmFlagS);
toc
%% ITK texture
cerrTestDir = getCERRPath;
cerrTestDir(end) = [];
if ispc
slashType = '\';
else
slashType = '/';
end
slashV = strfind(cerrTestDir, slashType);
cerrTestDir = cerrTestDir(1:slashV(end)-1);
cerrTestDir = fullfile(cerrTestDir,'Unit_Testing','tests_for_cerr');
globalGlcmDir = fullfile(cerrTestDir,'GlobalGlcmFeatures','win7');
resolution = [15 15 15]; % dummy resolution.
offset = [0 0 0];
scanFileName = fullfile(cerrTestDir,'mhaData','test1.mha');
maskFileName = fullfile(cerrTestDir,'mhaData','testMask1.mha');
test1M = permute(testM, [2 1 3]); % required to match coordinate system ...
%... between CERR and DICOm
test1M = flipdim(test1M,3);
mask1M = permute(maskBoundingBox3M, [2 1 3]);
mask1M = flipdim(mask1M,3);
delete(scanFileName)
delete(maskFileName)
writemetaimagefile(scanFileName, (test1M), resolution, offset)
writemetaimagefile(maskFileName, mask1M, resolution, offset)
% run ITK's textutre calculation
cd(globalGlcmDir)
tic
% system([fullfile(globalGlcmDir,'GlobalGlcmFeatures'), ' ', scanFileName, ' 4 1 4 ', maskFileName])
system([fullfile(globalGlcmDir,'GlobalGlcmFeatures'), ' ', scanFileName, ' ',...
num2str(nL), ' ', num2str(1), ' ', num2str(nL), ' ', maskFileName])
toc
fileC = file2cell('GlobalGlCMfeatures.txt');
% feature order from ITK:
% Energy, Entropy, Correlation, InverseDifferenceMoment, Inertia, ClusterShade
% ClusterProminence, HaralickCorrelation
featuresM = [];
for off = 1:size(offsetsM,1)
featuresM(off,:) = strread(fileC{off*2});
end
orderV = [11, 13, 10, 12, 5, 8, 9, 7, 6, 4, 1, 2, 3]; % map direction order ...
% ... between CERR and ITK.
maxEnergyDiff = max(abs((featuresM(orderV,1) - featureS.energy')./featureS.energy'*100));
maxEntropyDiff = max(abs((featuresM(orderV,2) - featureS.jointEntropy')./featureS.jointEntropy'*100));
maxInvDiffMomDiff = max(abs((featuresM(orderV,4) - featureS.invDiffMom')./featureS.invDiffMom'*100));
maxContrastDiff = max(abs((featuresM(orderV,5) - featureS.contrast')./featureS.contrast'*100));
maxClustShadeDiff = max(abs((featuresM(orderV,6) - featureS.clustShade')./featureS.clustShade'*100));
maxClustPromDiff = max(abs((featuresM(orderV,7) - featureS.clustPromin')./featureS.clustPromin'*100));
disp('========= Maximum difference between features for all 13 directions ==========')
disp(['Energy: ', sprintf('%0.1e',maxEnergyDiff), ' %'])
disp(['Entropy: ', sprintf('%0.1e',maxEntropyDiff), ' %'])
disp(['Inverse Difference Moment: ', sprintf('%0.1e',maxInvDiffMomDiff), ' %'])
disp(['Contrast: ', sprintf('%0.1e',maxContrastDiff), ' %'])
disp(['Cluster Shade: ', sprintf('%0.1e',maxClustShadeDiff), ' %'])
disp(['Cluster Prominance: ', sprintf('%0.1e',maxClustPromDiff), ' %'])