I think there's a very obvious error in the calculation:

The author jumps from 30mS increases the risk of cancer by 1/660 to 1 in 660 get cancer after the procedure.