1,743 karma · joined December 29, 2010
You can then play around with objects in 3D, 4D, and beyond:
https://enkimute.github.io/ganja.js/examples/coffeeshop.html...
Here's a fringe idea in the worse case: Connect one ECMO oxygenator to multiple patients at once using Type-O negative blood. You'd have to filter out the O-negative RBC's as they come in.
I'm a host in Seattle. The city now requires short-term rental operators to register as their own businesses.* If I'm a contractor for AirBnb, this makes sense.
However, if I'm an employee of AirBnb, then the Washington State income tax law would prohibit Seattle from collecting taxes on short-term operators.
Note that 30+ day stays are not* required to register, which doesn't make any sense since you're still operating a business.
I'm a resident at UW in Seattle. Despite being located in a wealthy city, UW mostly serves the poorer populations of the WWAMI states and consequently doesn't make a lot of money. The University has historically done everything to avoid raising residency salaries.
It's an interesting read about the steps residents had to get to unionize: https://uwha.com/history/
UW Medicine itself is not doing well financially (they lost $75 million last year): https://www.washington.edu/regents/files/2018/01/2018-02-B-2...
Harborview was the only profitable hospital in the system. Seattle is the second-largest tech city in the nation and housing prices have grown astronomically, just like in SF. The difference between UW and, say, Stanford or UCSF, is that UW's patient population comes from the WWAMI states. They don't typically treat the young tech works, although I have had a couple older Boeing/Microsoft patients. UW/Harborview patients continue to be mostly low-income Seattlites and tertiary care/trauma patients from the WWAMI states. The UW takes care of poor/rural patients while existing in a wealthy city. It's a unique place to work.
EDIT: I asked my fellow radiology residents about this event
"We had a nuc med lecture on the event! It's super interesting how they managed it."
"We had a separate nuclear medicine lecture on a Tuesday by the woman who helped managed the incident and is responsible for nuclear accidents."
"Was on HMC call that night. Physics lecture on that was useful. Radiation --> reflex call radiologist is a real thing."
"Cool. Honestly they should just reflex call them. I just paged them anyway."
“3.6 roentgens per hour. Not great, not terrible.”
https://papers.nips.cc/paper/4824-imagenet-classification-wi...
What you're saying is true, it should have worked in theory, but it just wasn't working for decades. The AlexNet team made several critical optimizations to get it work: (a) big network, (b) training on GPU, and (c) using a ReLU instead of tanh(x).
In the end, it was the hardware that made it possible, but up until their paper it really wasn't for sure. A good analogy is the invention of the airplane. You can speculate all you want about the curvature of a bird's wing and lift, but until you actual build a wing that flies, it's all speculation.
He glosses over amphetamines. They are highly effective at weight loss. So much so that ADHD patients have the stereotypical "Adderall skinniness" after 1-2 years of use. Addiction is a problem with stimulants, but psychiatrist routinely assess for addiction in the ADHD population. Why not prescribe Adderall for weight loss and similarly monitor those patients?
Which is better? A BMI of 40+ or a BMI of 25 and taking a daily stimulant?
The FDA agrees with the latter in at least one case. In 2015, the FDA approved the amphetamine lisdexamfetamine (brand name Vyvanse) for the treatment of binge-eating disorder.
However, my attitude towards radiology AI startups has changed quite a bit since starting residency. There really isn't a sustainable business model that involves selling standalone AI products, or even software suites. I think AI is a feature for the scanner, the PACS or for existing dictation software. That is, I think PowerScribe or GE would help a lot of radiologists by integrating AI features into their existing tech stacks.
I also think the only feasible exit strategy for radiology AI/ML startups is to get bought out for their tech talent and algorithms. I don't know of any hospitals/clinics/imaging centers that are buying AI products directly from startups. There's a ton of hype, but I'm not aware of a single positive-cash-flow radiology AI company.
“The bareheaded man needed a hat” falls squarely in the domain of contextually-aware models and would likely be transcribed correctly using deep learning.
Simple website for sending reminders. It didn't make any money for 4 years until I added commonly-requested features (templates, recurring reminders, and variables for recipient names). Now I have about 60 paying customers and it makes about $600 per month after all expenses paid. It's been running essentially autonomously for the past five months.
The iOS app is still difficult to use. I'm slowly rebuilding it. I'd say the main claim to fame at this point is if you Google "text message reminders" it's the first or second result.
Now I'm a radiology resident and have much more time.
Google: "There is a multiloculated fluid collection adjacent to the liver."
AssemblyAI: "There is a multi located fluid collection adjacent to the liver."
Google does better out of the box, but it isn't customizable (e.g. to a medical vocabulary). I think this is a major limitation of Google's approach, but certainly something they might offer in the future.
I do think AssemblyAI needs to offer a real-time dictation as this is the major need in medicine.
I don't agree with racial quotas, but there has to be some form of normalization for students whose parents/culture/childhood environment didn't value education.
Diet coke has no sugar, no fat, and less than 100 mg sodium. The specific results from Nature are about trehalose, a sugar not found in diet coke. I'm not aware of any diet soda that contains it, either. I realize it's an opinion piece, but pseudoscience and speculative articles should not be the domain of NYT.
Of note, the zero-calorie definition of artificial sweeteners is based on the amount excreted in urine/stool: It should be very close to the ingested amount. I find it hard to believe the conclusion drawn by the authors that other sweeteners (sucralose and saccharin), are consumed by bacteria since the amount ingested matches the amount excreted.