[breytex]

Shouldn't the end goal be just to train an ai on all the pdfs and give the doctors an interface to plug in all the details and get a treatment plan generated by that ai?

Working on the data structure feels like an intermediate solution on the way to that ai which is not really necessary. Or am I missing something?

[prmoustache]

I am not sure patients and doctors are interested in adding hallucination generators to the list of their problems.

[ska]

AI/ML techniques in medicine have been applied clinically since at least the 90s. Part of the reason you don't see them used ubiquitously is a combination of a) it hasn't worked all that well in many scenarios so far and b) medicine is by nature quite conservative, for a mix of good and not so good reasons.

[nradov]

What you're missing is there is no evidence that using AI models results in better patient outcomes than simpler clinical decision support tools.

[fl0id]

Your end goal maybe. Not patients or doctors goal for sure.

[pjc50]

How does your treatment AI get its liability insurance?