haldujai

17 ms·

181.
by haldujai 3y ago
It's not because of pump issues but more glucagon stability and secondarily whether bihormonal is clinically advantageous. Tandem is working on a dual chambered pump. https://diatribe.org/jdrf-and-tandem-diabetes-care-a
182.
by haldujai 3y ago
> In any case you will still need some a-priori info about meals and planned activities Not necessarily, at least not via patient input. In the albeit small Inreda studies manual announcement of exercise and meals wasn't required (
183.
by haldujai 3y ago
No, there isn't a hardware problem and such systems exist. Several trials[1-3] dating back several years have looked at "bionic pancreas" or a closed loop bihormonal system. Inreda (a Dutch company) has a CE-marked device[4]
184.
by haldujai 3y ago
Absolutely no one is getting a bilateral mastectomy or chemotherapy for DCIS. This is by far the most inaccurate medical claim I've ever seen on HN. Where on earth did you get this from? The whole point of diagnosing DCIS on screening
185.
by haldujai 3y ago
It's unequivocal that screening reduces breast-cancer specific mortality. None of the studies are powered enough to draw a positive or negative conclusion for overall mortality. The second last link is an explanation of why this is the
186.
by haldujai 3y ago
Read all of the caveats. None of this is still relevant. Cancer-specific mortality is definitely reduced. There is no study powered enough to draw conclusions on all-cause mortality, which may not be the best measure anyway.
187.
by haldujai 3y ago
I don't know why you're downvoted. The commenters below are misunderstanding things, the NNT page is old and even then it's being mis-cited. I won't go into it but read the caveats section for the author's own expla
188.
by haldujai 3y ago
That's not the criteria for population level interventions. This is to answer the question of whether should the USPSTF make a positive recommendation and force insurers to pay or if public health systems should adopt this. You're
189.
by haldujai 3y ago
Cardiac cirrhosis is just a bad example, it’s difficult to estimate prevalence but it’s very rare and even then most cases don’t matter as it reflects a state of severe systemic dysfunction (i.e. patient is on deaths door for several other
190.
by haldujai 3y ago
Not sure what the relevance of cardiac cirrhosis is to this discussion? You’re talking about an uncommon end stage complication of chronic severe right heart failure. To the parent comment’s question: yes, prevention is better. If you preve
191.
by haldujai 3y ago
Fully agree that we still need to develop treatments but from a prevention perspective if you maintain a healthy weight and diet (i.e. avoid metabolic syndrome) don’t smoke, and don’t use cocaine you should be doing really well from a coron
192.
by haldujai 3y ago
That future was the same one advertised 13 years ago when I started medical school. We were already storing cord blood back then. I’m not saying it will never happen, but this is a really challenging treatment paradigm to get right and I do
193.
by haldujai 3y ago
> We are already there. The FDA has approved hematopoietic stem cell transplantation. Source? To my knowledge what’s approved is not iPSC. Hematopoietic progenitors are a different challenge and “low hanging fruit”. https://ww
194.
by haldujai 3y ago
You would probably want to induce differentiation in vitro before injecting to make sure it turns into the right cell type but stem cell therapy hasn’t worked out particularly well. One of the easiest problems to understand is that the cell
195.
by haldujai 3y ago
We have been “close” to some breakthrough with stem cells for at least 15 years now except every time it either doesn’t work or ends up causing cancer (e.g. tumorigenicity and immunogenicity). I’ll believe it when I see it.
196.
by haldujai 3y ago
It’s not as simple as you’re saying but you’re also a bit off for the record. Annual net migration per 1000: Of all countries the US ranks #39. Of developed countries with a population > 10M the US ranks #8 (#9 if you include Saudi Arabi
197.
by haldujai 3y ago
I assume you’re not trying to be intentionally misleading but those graphs do not support your assertion that the US, presumably referring to the average American, is less wealthy today. You can’t just cherry pick things that have become mo
198.
by haldujai 3y ago
Except if your leg is blown off every hospital in the US is legally required to treat you whether or not you have insurance or even if you’re homeless, so Gary is not your only option.
199.
by haldujai 3y ago
You’re right that the vast majority of the time it’s low effort but it’s hard to identify those cases with high specificity up front. The issue in medicine is there is huge class imbalance, 90%+ of encounters are essentially “negative” or n
200.
by haldujai 3y ago
“Bullshit” is almost always a post hoc diagnosis. In the last 2 years I’ve seen 3 20-somethings with a few days of flu symptoms having used a virtual walk in clinic service before coming to the ER (this is Canada where primary care is mostl
201.
by haldujai 3y ago
Zero chance with regulation. We still don’t let machines final read ECGs and insurance companies pay a physician (typically a cardiologist) to overread them. These have been FDA approved algorithms for over a decade already.
202.
by haldujai 3y ago
It would be interesting to see what the breakdown is on H100/A100 users. I would expect most inference users are similar to my lab and max out at a DGX node rather than being the bulk of users. PyTorch has gotten a lot better on TPUs t
203.
by haldujai 3y ago
> I also don’t know any professional devs who have used tools like copilot and said they were anything but a toy. Really? I'm academic now but I find Copilot at least moderately helpful when I'm writing a library. It's pre
204.
by haldujai 3y ago
There may also be a third more important concern: AI/LLM generated works are not generally copyrightable.[1] The Copilot lawsuit should answer concern #1 more definitively. #2 is already solved by running your own model or using Azure
205.
by haldujai 3y ago
That seems like a likely explanation, probably won't get into legal trouble for using an OpenAI model for a research paper but redistributing said model may be upsetting enough for OpenAI trigger a legal challenge. Unnatural language u
206.
by haldujai 3y ago
I thought this specific model was referring to self-instruction using both synthetic prompts (generated from few-shot in-context prompting of presumably some OpenAI model, the original paper used text-davinci-002) as well as synthetic code
207.
by haldujai 3y ago
> You seem to be moving the goalposts. First it was using in production, now it's building foundation models.. I'm not moving the goal posts but maybe I wasn't clear. When we talk about corporations adopting fine-tuning an
208.
by haldujai 3y ago
> Customer support. I've never seen so many people get up to speed so fast on something. It's unlike anything ever. Are you saying that you've seen enterprises developing, training and running their own in-house LLMs from
209.
by haldujai 3y ago
> Large enterprises are already putting them into production. I have direct experience with it. For what function have you experienced LLMs being used at significant scale in production right now? It seems unlikely most enterprises have
210.
by haldujai 3y ago
> I would say Intel likely keeps pushing, but I'm less convinced they can actually make much headway in the near future. It seems that Intel is making great headway on their fabs and may somehow pull off 5 nodes in 4 years. Intel 3

More ›