[whimsicalism]

Nonsense.

The medical possibilities that will be unlocked by large generative deep multimodal models are on an entirely different scale from "statistical diagnoses." Imagine feeding in an MRI image, asking if this person has cancer, and then asking the model to point out why it thinks the person has cancer. That will be possible within a few years at most. The regulatory challenges will be surmounted eventually once it becomes exceedingly obvious in other countries how impactful this technology is.

[DrScientist]

But in your scenario - which part is adding the value?

Your deep multimodal models or the MRI imaging?

What you are essentially saying is the signal is so subtle that only a large NN can reliably extract it.

While that may well be the case, it would be better to have a scan/diagnostic that doesn't need that level of signal processing to interpret.

For example - you don't need a large generative deep multimodal model to read a Covid antigen or PCR test.

[whimsicalism]

There are tons & tons of conditions that do not have easy scans/diagnostic and rely on subtle signals - especially if they are not a binary yes/no but a regression style prediction.

We've picked a lot of the low-hanging simple to extract signals, we need large models to go to the next phase for things like parkinsons, etc.

[DrScientist]

I'm not saying there isn't stuff that can't be done more reliably - but I'd argue long term might be better investing in getting better data - rather than better fishing in a pool of low quality data.