[graeham]

To be truely useful as a diagnostic in major decision-making, it needs to be 90-95%+ accurate.

But unfortunately there isn't anything close to this yet.

High blood pressure, for example, probably more accurate than the retinal image technique here. https://www.ncbi.nlm.nih.gov/pubmed/25632496

The major goal here would be to screen for people to go for a more advanced test, like an MRI. But even that wouldn't hit the 90-95% level of predictor.