I agree with you, except on the timescale. I'd say next 100 years. For your prediction to be right, we'd need a pretty good prototype of the 'brain cap/scanner' in about 20 years. I don't know if that's enough to be able to do this type of diagnosis, based of course on what we currently know.