number of unaggregated class C's in swamp?

How would one go about determining how many /24 /23 ... /19 unaggregated
advertisements are in the swamp. I assume it would have to be examined
by AS, and look for multiple paths to find multi-homed, but it should be
doable with the current database, shouldn't it?

Has someone already done this? I asked Merit yesterday, and they don't
seem to have the tools.
          Key fingerprint = 2E 07 23 03 C5 62 70 D3 59 B1 4F 5E 1D C2 C1 A2