dontreact

5 ms·

211.
by dontreact 4y ago
Yes, it's not a 50% chance of false positive. It's 50% chance of false, given positive test.
212.
by dontreact 4y ago
No one just treats a mammography or tomography positive without other follow up tests. It's not the standard of care. The next step is to get an ultrasound and then a biopsy.
213.
by dontreact 4y ago
https://pubmed.ncbi.nlm.nih.gov/33141657/ "Long-term follow-up of the UK Age RCT" "There was a statistically significant 25% reduction in mortality from breast cancers diagnosed during the intervention p
214.
by dontreact 4y ago
It will take a long time to sort this out. I think that eventually, the ability for AI to review 3D in greater detail will improve on false positives and false negatives. For CT vs. XRay there have already been 2 long term randomized contro
215.
by dontreact 4y ago
You don't, you just measure long term outcomes like mortality in a treatment and control arm. Potentially you have to use a synthetic control. I am most familiar with NLST which actually was a long term randomized trial.
216.
by dontreact 4y ago
Yes exactly. No one is going to put you on chemo because of a mammogram. There are several followup tests including ultrasound and biopsy.
217.
by dontreact 4y ago
This is all certainly true but it's important to look at the bigger picture. What is the overall impact of catching cancer earlier and saving lives? That is enormous as well. One way this is sometimes talked about in public health lite
218.
by dontreact 4y ago
It's good to be publishing stats like this so that the psychological harm of false positives is lower. It's scary to get a false positive. However, in my opinion, there has been a systemic resistance to screening because of statis
219.
by dontreact 4y ago
Yeah this is why in practice the followup to a mammogram is never going to be another mammogram. It's a different test which is more invasive but also produces far less false positives: taking a tissue sample.
220.
by dontreact 4y ago
However, despite the high false positive rate, breast cancer screening is a net beneficial procedure (though not without controversy) in that when applied in a population, long term outcomes like mortality improve. (edit yes of course cance
221.
by dontreact 4y ago
Yes exactly. In fact mammograms are already double-read and agreement is quite high, so this is definitely one of those types of tests. There are benign masses that are radiologically impossible to distinguish until you get a biopsy.
222.
by dontreact 4y ago
This is actually a really cool idea. Looking forward to see how/if it scales up to humans. Would be a sad but hilarious example of why lab bench testing and testing in humans or animals is such a large gap Bench: The worms swim towards
223.
by dontreact 5y ago
Probably not a good idea for Google to go after other companies in this way given all the antitrust specter. NYTimes headline “Google abusing monopoly position to muscle out delivery startups”
224.
by dontreact 5y ago
I thought it was Grubhub/Doordash actually taking the orders, Google is just providing a link to their service.
225.
by dontreact 5y ago
Funny how there is pushback on the YCombinator forum to an attempt to hold accountable a YCombinator company that actually has a direct hand in doing this to restaurants. How would Google know which postings are fraudulent and which aren&#x
226.
by dontreact 5y ago
So then the hospital or lab are under HIPAA, but not the forensic lab.
227.
by dontreact 5y ago
A forensic lab doing things with DNA may have no business relationship whatsoever with any covered entity (in fact I would guess that this is the most likely case).
228.
by dontreact 5y ago
Yes but my sense it is coming pretty soon for ES, and switching to Pinecone has a huge rewrite cost. As NNs keep getting better, vector search keeps becoming more important, and it becomes more and more existential for elasticsearch to do t
229.
by dontreact 5y ago
The way that vector indices work typically can make doing CRUD with them a real challenge. There is definitely novelty in being able to do both ANN indexing and fast high throughput CRUD. In addition, the R of crud is hard to combine with v
230.
by dontreact 5y ago
I pay 25$ a month to Textedly to enable me to easily send text messages to a group of people and enable signups to a list via a short code.
231.
by dontreact 5y ago
I agree with you, but I also see violence as an inevitable consequence of growing inequality. If we continue to let wealth inequality grow, more and more people will feel the game is unfair, and choose to play another game, even if the &quo
232.
by dontreact 5y ago
I saw a presentation at a conference that pretty strongly suggested they will have a remote driver take over at certain points (the car will assess it can’t continue and remote driver takes over)
233.
by dontreact 5y ago
Interesting how much hate I have seen for how Tesla didn’t do this before and how little enthusiasm or support there seems to be for this. This seems great!
234.
by dontreact 5y ago
No, you have to pay 11k to publish
235.
by dontreact 5y ago
If pelosi were murdered (and potentially other senators too) Do you think the senate would still have certified the election? Or would Trump still be president? I think the fear is that the mob was almost able to make it so that Trump’s lie
236.
by dontreact 5y ago
What do you think would have happened if things went a little differently and the mob got to Nancy Pelosi? They were very close to doing so.
237.
by dontreact 5y ago
I totally agree that culture is a big factor. I just think the conversation in the US around healthcare is kind of limited to pro or anti “Medicare for all” when there are actually many ways of doing things which do not require a single pay
238.
by dontreact 5y ago
The Japanese system is not a single payer healthcare system and in terms of outcomes vs. cost is one of the best in the world, so I am not so sure that single-payer is the most reasonable solution. Particularly in the US where voters are av
239.
by dontreact 5y ago
Yes but if the game you’re playing will only hand you losses time and again, then you will look for a different game, or lose motivation/give up.
240.
by dontreact 5y ago
I know this from working there, but it’s kind of a shameful “secret” as far as I can tell?

More ›