this post was submitted on 30 Jun 2025
0 points (NaN% liked)

Technology

207 readers
4 users here now

Share interesting Technology news and links.

Rules:

  1. No paywalled sites at all.
  2. News articles has to be recent, not older than 2 weeks (14 days).
  3. No videos.
  4. Post only direct links.

To encourage more original sources and keep this space commercial free as much as I could, the following websites are Blacklisted:

More sites will be added to the blacklist as needed.

Encouraged:

founded 1 month ago
MODERATORS
 

The Microsoft AI team shares research that demonstrates how AI can sequentially investigate and solve medicine’s most complex diagnostic challenges—cases that expert physicians struggle to answer.

Benchmarked against real-world case records published each week in the New England Journal of Medicine, we show that the Microsoft AI Diagnostic Orchestrator (MAI-DxO) correctly diagnoses up to 85% of NEJM case proceedings, a rate more than four times higher than a group of experienced physicians. MAI-DxO also gets to the correct diagnosis more cost-effectively than physicians.

top 3 comments
sorted by: hot top controversial new old
[–] [email protected] 0 points 23 hours ago

Seems kind of worrying that the best doctors in whatever field of medicine this is, can only correctly diagnose cases no better (or very slightly better) than 1 out of 5 times.

[–] [email protected] 0 points 23 hours ago (1 children)

I know that I might be the only Lemmy user happy with this, but AI applications in the medical field seems very promising for lowering costs and being more accurate.

[–] [email protected] 0 points 21 hours ago

Instead of lowering costs, remove capitalism and make doctors actually care about health rather than their paycheck and everything will work much better than a stupid AI.

As for accuracy, that's bullshit, you cannot measure medical success through the metric of valid diagnostics. A doctor is supposed to give a diagnosis that is reasonable based on a context ; an AI will try to find the answer that seems the most "likely" according to some data without any understanding of the context or risks.

This shit is really the same crap that happened with people eating radioactive crap thinking it would make them stay healthy, simply because it was the new cool tech. The difference being that radioactivity is a physical concept independent from human bullshit, while AIs are made by stupid humans, for stupid users, using human stupidity as its only source of information. It exacerbates the worst of humanity and frankly, isn't developed enough (and society isn't developed enough) to have a real use.