joshgel

12 ms·

91.
by joshgel 6y ago
I'm trying to learn more about this. What I am reading suggests this is the wrong frame. What if all the money spent over-policing these neighborhoods was instead spent on social services to support the people living there? Would that
92.
by joshgel 6y ago
88% is definitely not true. That paper retracted that claim.
93.
by joshgel 6y ago
Can I just add that seasonal influenza varies widely from year to year? [0] It just so happens that this year was a moderately bad year in the US. I don't know about China. But, they'd have to exclude differences from a bad flu
94.
by joshgel 6y ago
There are a number of visual examples that help explain the difference between equality, equity and justice. But this one I particularly like: https://www.reddit.com/r/coolguides/comments/gq2n31/inequali.
95.
by joshgel 6y ago
Does it run iMessage on Windows?
96.
Observational Study of Hydroxychloroquine in Covid-19 finds no benefit(nejm.org)
2 points by joshgel 6y ago|0 comments
97.
by joshgel 6y ago
Any good takes on the benefit/harm ratio of these pre print servers? Yes, it’s good to get info out quickly, especially in a pandemic. Yes, it’s very important that this kind of science is not stuck behind paywalls. But I really worry
98.
by joshgel 6y ago
Also see: https://github.com/smarterdx/covid-19 Similar concept with different data that we are deploying in US hospitals. Using React and Hasura.
99.
by joshgel 6y ago
But they didn't even check the Vitamin D levels. They only checked patients that had Vitamin D levels checked by physicians. So presumably this was a different population that was more sick by definition, because their physicans were c
100.
by joshgel 6y ago
We aren’t treating oxygen levels. Higher oxygen levels now that result in lower oxygen levels later might actually be a bad thing. I have no doubt that proning increases the readings we are getting on pulse ox monitors, sometimes pretty imm
101.
by joshgel 6y ago
That 88% number is pretty misleading. They only looked at: (number of patients who died) / (number of patients who died + number of patients discharged alive). And didn't consider the other patients on ventilators who are still al
102.
by joshgel 6y ago
Even better would be randomized experimental data that this was better than an alternative... 'It works' can mean lots of different things...
103.
by joshgel 6y ago
Our* guidance has been HFNC only in negative pressure rooms because of the aersolization risk. We've converted a huge number of beds to negative pressure in order to maximize use of HFNC. Anecdotally, I feel like it works, but would be
104.
by joshgel 6y ago
And also, if you are intubating the person, they are probably pretty close to death. And then on top of that you paralyze them (with a medication, temporarily). So you are stopping their breathing. You have a few seconds in these people to
105.
by joshgel 6y ago
HANG ON! This only considers patients already discharged or died at the end of the study some ~300 patients. But there are still ~1000 patients who were on mechanical ventilation that are still in the hospital (so aren't included in th
106.
by joshgel 7y ago
Also a plug for https://nbdev.fast.ai/ which lets you run tests, produce documentation and version control notebooks.
107.
by joshgel 7y ago
I spent the weekend in the Emergency Dept of a major NYC hospital. In the past 3 weeks I've probably looked at ~250 Chest X-rays of COVID-19+ or likely patients. (Not getting many CT scans currently). You really don't need AI to f
108.
by joshgel 7y ago
Co-founder of SmarterDx here. SmarterDx is releasing a tool that will help doctors and hospitals more rapidly review the relevant data for their COVID patients. We hope this will enable better triage, more rapid review and the ability to id
109.
Show HN: Covid-19 Patient Review Tool for Hospitals Open Sourced(github.com)
20 points by joshgel 7y ago|1 comments
110.
by joshgel 7y ago
As a grandchild of a holocaust survivor, this was told to me repeatedly.
111.
by joshgel 7y ago
That’s not the only problem. Say you aren’t high risk, great! But do you really want to live in a world without a functioning health care system. Because that is what you are promoting. When hospitals get overwhelmed, even minor issues can
112.
by joshgel 7y ago
I’m not sure about the exact mechanics, but we are being advised to minimize ‘bagging’ (ie using the ambu-bag to help patients breath before they are intubated) as much as possible. We worry about the markedly increased risk of aersolizatio
113.
by joshgel 7y ago
We literally do this for hospital rooms. There are uv devices that the cleaners bring in and turn on for 30-45 min. So good idea!
114.
by joshgel 7y ago
This isn’t really true. Stents in patients with heart attacks have a clear mortality benefit except in certain high risk situations when bypass surgery has better outcomes. Stents are quite routine and much, much, much easier to do than byp
115.
by joshgel 7y ago
I think part of it, but also not the entire story. Take cancer, for example. The treatment is so toxic (potentially deadly) that we want to make sure that our diagnosis is as close to 100% as possible. To the point where if you want to tran
116.
by joshgel 7y ago
Can confirm the thrust of this story. And I trained in the last 10 years in NYC. Modern healthcare is little different.
117.
by joshgel 7y ago
Same, my hospital uses Tableau, but expensive and some what clunky. This could be much easier and faster.
118.
by joshgel 7y ago
Traditional chemotherapy is toxic to all cells. Cancer cells grow rapidly (part of what makes them dangerous). So the toxic effect gets the cancer cells first, since cells that are growing take up the most chemotherapy. But other cells grow
119.
by joshgel 7y ago
Can I just throw a big dose of caution here? This is a review study, looking at the available literature of this technique on cancer. They found a total of FIVE studies with 88 total human participants across all 5 studies. We, the medical
120.
by joshgel 7y ago
I also agree, but I'm a physician (an internist => generalist). As I grow in my career, I'm finding that when I encounter a new disease or procedure or treatment and go to read about it I learn an order of magnitude faster than

More ›