Adapting the gTruthtoXY helper function

I'm currently trying to implement the "Convert Image Labeler Polygons to Labeled Blocked Image for Semantic Segmentation" tutorial to solve an issue generating masks for hand MRIs (DICOM images) in which polygons have been used to label each individual bones with each of the 10 bones labelled being considered a seperate class

I need to create masks indicating the position for all bones to train a DL network to segment this automatically. However I'm having a hard time with it considering I've used polygons and I ideally need pixel classification labels. The tutorial seems to outline a way to get the output that I need but I need to modify the help function (gTruthtoXY) to include all 10 classes of bones within the mask. The current helper function only includes 2. How would I go about modifying this helper function?

Supporting Function

gTruthtoXY converts the polygon ROI coordinates and label data stored in the table labelData into cell arrays suitable for input into the polyToBlockedImage function.

function [roiPositions,roiLabels] = gTruthtoXY(labelData)

totalROIs = numel(labelData{1,1}) + numel(labelData{1,2}{:});

roiPositions = cell(totalROIs,1);

roiLabels = zeros(totalROIs,1);

% Obtain label names from the labelData table

labelName = labelData.Properties.VariableNames;

roiIdx = 1; % Initialize ROI index

% Loop through all labels

% Assign a numeric label of 2 to tumor tissue; 1 for normal tissue

for j = 1:numel(labelData)

% All ROIs for a given label

data = labelData{1,j}{:};

if(isequal(labelName{j},"tumor"))

for k = 1:numel(data)

roiLabels(roiIdx) = 2;

roiPositions{roiIdx} = data{k};

roiIdx = roiIdx + 1;

end

else

% For other ROI labels

roiLabels(roiIdx) = 1;

roiPositions{roiIdx} = data;

roiIdx = roiIdx + 1;

end

end

end

Any help much appreciated! Thanks

Answer 1

[Yuvraj Singh]

Hi Lily,

To achieve the expected result.

Update the calculation of "totalROIs”

You would have to add ROIs for different labels. 

%Calculating total ROIs if tumor_type_n is a cell type 
totalROIs_tumor_type_n = numel(labelData{1,n}{:}); 
%Calculating total ROIs if tumor_type_n is a double vector 
totalROIs_tumor_type_m = numel(labelData{1,m}); 

totalROIs = totalROIs_tumor_type_1 + totalROIs_tumor_type_2 ....  
    + totalROIs_tumor_type_10 

Add additional condition in for loop for different label types.

if(isequal(labelName{j},"tumor_type_1")) 
    
elseif(isequal(labelName{j},"tumor_type_2")) 
    . 
    . 
    . 
    . 
    . 
    . 
    
elseif(isequal(labelName{j},"tumor_type_9")) 
    
else  
    
end  

Add labelling code in the if condition. It varies if label is a double vector or a cell.

%Code inside if condition for tumor_type_n if it is cell type 
for k = 1:numel(data) 
    roiLabels(roiIdx) = n; 
    roiPositions{roiIdx} = data{k}; 
    roiIdx = roiIdx + 1; 
end 
%Code inside if condition for tumor_type_m if it is a double vector 
roiLabels(roiIdx) = m; 
roiPositions{roiIdx} = data; 
roiIdx = roiIdx + 1; 

[Lilly]

Thank you so much for your answer, I'll implement this and let you know how it goes