It's been done a lot of times, with wildly different results. The problem is that to get a final number like that, you have to make some absolutely wild guesses with almost no meaningful information to estimate, such as the probability of the first cell developing on a planet with absolutely ideal conditions for it to happen. We can't even really say that we even have a single data point on this, because we don't even know necessarily how ideal the conditions on earth were during that time, or how many times it happened but never replicated. That's not even touching on the more lofty questions like, are cells even necessary for life? Depending on what numbers you pull out of thin air for these questions, you could come up with a million civilizations in the Milky Way, or almost none in the entire universe. That's the really scary part: we don't even know what we don't know.