I am stuck on some non-coherent fault tree analysis. I have a non-coherent fault tree for which the TOP event breaks down to TOP = AD' + DA' + A'E. These are (I think) some of the prime implicants of the fault tree. There is also another prime implicant ED'. I've been trying to work through it with De Morgan's laws but with no luck getting that final prime implicant (ED'). I've attached an image of the fault tree. Any ideas?