How to find the sensitivity and specificity from the confusion matrix?

[madhan ravi]

upload your code , it's hard to debug screenshots

[Win Sheng Liew]

global net imageSize featureLayer classifier

f = waitbar(0,'Please wait...');

rootFolder = fullfile('c:\', 'Users', 'Ws Liew', 'Documents', 'MATLAB', 'Dataset'); categories = {'normal', 'abnormal'}; imds = imageDatastore(fullfile(rootFolder, categories), 'LabelSource','foldernames');

tb1 = countEachLabel(imds)

minSetCount = 0.6;

imds = splitEachLabel(imds, minSetCount, 'randomize');

normal = find(imds.Labels == 'normal', 1); abnormal = find(imds.Labels == 'abnormal', 1);

str = {'Select a CNN Architectures: '}; list = {'GoogLeNet','AlexNet','ResNet','VGG-16','VGG-19'}; [indx,tf] = listdlg('PromptString', str, 'ListSize', [200 130], 'SelectionMode', 'single', 'ListString', list);

waitbar(.25,f,'Loading Pre-trained Network');

if indx == 1 net = googlenet(); featureLayer = 'loss3-classifier';

elseif indx == 2 net = alexnet(); featureLayer = 'fc8';

elseif indx == 3 net = resnet50(); featureLayer = 'fc1000';

elseif indx == 4 net = vgg16(); featureLayer = 'fc8';

else indx == 5 net = vgg19(); featureLayer = 'fc8';

end

net.Layers

net.Layers(1)

net.Layers(end)

numel(net.Layers(end).ClassNames)

waitbar(.4,f,'Training and Testing Image Datasets');

[trainingSet, testSet] = splitEachLabel(imds, 0.3, 'randomize');

imageSize = net.Layers(1).InputSize; augmentedTrainingSet = augmentedImageDatastore(imageSize, trainingSet, 'ColorPreprocessing', 'gray2rgb'); augmentedTestSet = augmentedImageDatastore(imageSize, testSet, 'ColorPreprocessing', 'gray2rgb');

w1 = net.Layers(2).Weights;

w1 = mat2gray(w1); w1 = imresize(w1,5);

trainingFeatures = activations(net, augmentedTrainingSet, featureLayer, 'MiniBatchSize', 32, 'OutputAs', 'columns');

trainingLabels = trainingSet.Labels;

waitbar(.65,f,'SVM Classifier Training');

classifier = fitcecoc(trainingFeatures, trainingLabels, 'Learners', 'Linear', 'Coding', 'onevsall', 'ObservationsIn', 'columns');

testFeatures = activations(net, augmentedTestSet, featureLayer, 'MiniBatchSize', 32, 'OutputAs', 'columns');

predictedLabels = predict(classifier, testFeatures, 'ObservationsIn', 'columns');

testLabels = testSet.Labels;

confMat = confusionmat(testLabels, predictedLabels);

confMat = bsxfun(@rdivide,confMat,sum(confMat,2))