Trim only edge NaN values

Say I have a variable:

X = [NaN, NaN, 2, 5, 1, 6, NaN, 1, NaN NaN];

I need to remove/trim NaNs only when they occur at the edges, either at the beginning or the end, so X would be:

X = [2, 5, 1, 6, NaN, 1];

where the NaN occurring at the middle was untouched. I wrote some codes that work, but I think the codes could be much simpler (I tend to overcomplicate my codes 😅). Please suggest a simpler way to do this. Thanks!

NaNGroups = regionprops(isnan(X), 'Area').Area;
if isnan(X(1))
    X(1:NaNGroups(1)) = [];
end
if isnan(X(end))
    X(end-NaNGroups(end)+1:end) = [];
end   

Answer 1

[KSSV]

X = [NaN, NaN, 2, 5, 1, 6, NaN, 1, NaN, NaN];
A = isnan(X) ;

idx = find(A & diff([0 A]) == 1 & diff([A 0]) == 0) ; 
idx = [idx(1) idx(1)+1 idx(end) idx(end)+1] ;
X(idx) =[]

Also you can use:

find(isnan(X),2,'first')
find(isnan(X),2,'last')

[Adib Yusof]

Thank you so much!