This extracts all digits and all '.' from a:
numstr = a(regexp(a, '[\d\.]'))
This extracts all numbers between ( ), where number must have at least one digit before the point and one digit after the point:
numstr = regexp(a, '\((\d+\.\d+)\)', 'tokens')
The outer \( and \) match the '(' and ')', resp., in a, and the inner ( ) are meta-characters to group a token.