bootstrapping on a large data set without loop

I have a vector a(k,1) and a 3-dimensional matrix b(m,n,k). I want to use bootstrapping to see if the mutual information between a and b based on the 3rd dimension is significant or not. I can use this loop: for i=1:m for j=1:n bootci(1000,@mutualinfo,a,b(i,j,:)); end end ,but it is time consuming, is there any way that I can perfrom it without loop?