Vectorized way to get unique strings of cell sub-arrays

I have an nx12 cell array where the first column itself is a 1x2 cell array containing a string. I was wondering, if there is a vectorized way to get the unique strings from each second column of my sub-array.

Either

unique(MyData{:,1}{1,2})

nor

cellfun(@unique,MyData{:,1}{1,2})

did work as it gives me the error 'Bad cell reference operation.'! I could use a loop to reorganize my data but I think there is a more easier approach to this which I just don't see right now.

Answer 1

[Guillaume]

If all the cell arrays in the first column are indeed all the same size, you can concatenate them into a single cell array, which is then easy to index:

firstcolumn = vertcat(MyData{:, 1});
unique(firstcolumn(:, 2))