News for hackers. Stuff that matters.
Re:Radiology specific foundation model released by Harrison.ai by rscho on Tuesday September 10, 2024 @08:08AM (#41499922)
In my experience, it doesn't seem to be a common occurrence. At least, I personally know no one doing that. YMMV, I guess. Also, it seems to me like a very bad idea to do that.
Re:Radiology specific foundation model released by Harrison.ai by delichon on Tuesday September 10, 2024 @09:14AM (#41500494)
As a type 2 diabetic I used a couple of different glucose monitors and got a lot of benefit from them. I gave one to a friend who I thought had diabetic symptoms. First I tried to get him to go to a doctor but he wouldn't. But he tried the CGM and found numbers well into the diabetic range. Then he immediately changed his diet and started treatment. Not sure but I may have committed a crime.

What was the potential harm that was greater than the reward?

Re:Radiology specific foundation model released by Harrison.ai by haldujai on Tuesday September 10, 2024 @10:21AM (#41501074)
Potential harm is always the same - misdiagnosis and/or mismanagement.

It’s probably very low in the context of CGM and diabetes as the potentially harmful treatments require prescriptions.

Device prescription requirements are usually due to product labelling and the manufacturers application. There are OTC fingerstick glucometers and CGMs approved.

Re:Radiology specific foundation model released by Harrison.ai by rscho on Tuesday September 10, 2024 @09:50AM (#41500742)
Accidentally diagnosing someone is quite different from someone healthy trying to 'optimize' their health, whatever that means...
Re:Radiology specific foundation model released by Harrison.ai by delichon on Tuesday September 10, 2024 @02:37PM (#41504074)
Strong disagree, diagnosing chronic dysfunction is essential to optimizing health. There's a big difference between an optimal blood glucose range and one that triggers insurance companies to authorize treatment. If you only pay attention to the later it could cost years of healthy life.

It's like, not being obese enough for your insurance company to pay for medical intervention doesn't mean that your weight is optimal enough to enjoy a long retirement.

Re:Radiology specific foundation model released by Harrison.ai by nradov on Tuesday September 10, 2024 @11:23PM (#41507747)
Bad idea how? It's expensive but not dangerous. Some people find the results interesting, and serious athletes have had some good results using them to optimize diet and training.