Check out our work published in EHJ today! This was a significant part of my dissertation and it was a joy to work with @PierreEliasMD and the rest of the team!
Today in #EHJ, we present the 1st time an AI model has outperformed radiologists in detecting cardiac pathology.
❤️🩹 Potential screen for #CHF w/ CXRs
🌁 Trained @ColumbiaMed validated @StanfordAIMI
🩻 Entire dataset (75k CXRs) released
@ESC_Journalsshorturl.at/cOT16
Can you detect Stage B heart failure from a chest x-ray? Our ML model learns to detect LVH/dilated LV from CXRs done within 1 year of echo. Great work with @sabhave/@aperotte/@NirUrielMD. Currently validating at Stanford thx to @David_Ouyang & @mattlungrenMD! #ACC21
India is in the throes of a horrendous COVID surge
Horrendous
They are struggling to get more people vaccinated
We are sitting on 35-40 million doses of Astra Zeneca vaccine Americans will never use
Can we please give or lend them to India? Like may be now?
It'll help. A lot
And further evidence that dispersed EMR - with no interoperability -- is a liability in a pandemic.. look at what the UK NHS database achieved in a few weeks. We should be ashamed ...
x.com/bengoldacre/status/125…
OUR NEW PAPER.
Factors associated with COVID-19-related hospital death in the linked electronic health records of 17 MILLION adult NHS patients.
Largest study of its kind ever, anywhere.
The power of UK / NHS data, realised.
opensafely.org/outputs/2020/…medrxiv.org/cgi/content/shor…
1/ A student just asked me why, if the p-value for a study is 0.04, we can't say the study has a 4% chance of being a false positive. First off, we definitely can't, even under idealized conditions - here's a brief thread as to why.
This vote will allow senators to show once and for all where everyone stands on #NetNeutrality. We hope that we can make this bipartisan & our GOP colleagues will join us in standing up on behalf of average consumers & middle class families. x.com/netflix/status/9938873…
I made this animation to show the avg share of cause of death from @CDCgov vs. avg share of @Google search trends and headlines from @guardian & @nytimes. The data was collected by Owen Shen et al. in a great post: bit.ly/ChartDeath
Source code: bit.ly/DataVizCode
Billions of lab tests are performed in the U.S. each year, and most are not stratified by demographic variables (e.g. race, sex, age). Should they be? John Ioannidis, @chiragjp, and I explore some of the thorny challenges in our new @JAMA_current piece: jamanetwork.com/journals/jam…
The Trump Administration’s new census question about citizenship could lower the response rate, affect accuracy & is a clear attempt to undermine representation in states like CA where the immigrant population is larger than in any other state. mercurynews.com/2018/03/27/2…