sanjatkanjilal.bsky.social
Assistant Prof @DeptPopMed @harvardmed | ID doc & Assoc Med Director of Micro @BrighamWomens | 🧐 machine learning for dx/tx/prevention of infectious diseases
26 posts
43 followers
47 following
Regular Contributor
Conversation Starter
comment in response to
post
Perfectly said! I’d only add we don’t want to FAFO during this unfortunate data gathering period. I suggest staying up to date with where measles is being detected locally and nationally and avoid those areas if you can
comment in response to
post
Stevil. Evupid.
comment in response to
post
I’m not a Luddite by any means but this article provides really bad advice.
comment in response to
post
Thanks to you and @angelahuttner.bsky.social for your leadership on this!
comment in response to
post
💯. Dark times ahead. Tough to see this as an ID doc.
comment in response to
post
Yea. But his years of anti-vax propaganda has already had an impact on vaccination rates; he is responsible for some of this tragedy
comment in response to
post
Nice to reconnect here Ilan! And thanks for the shout out
comment in response to
post
Very nice, thx. I explained the negative impact that med schools have for US healthcare and how it influences choice of specialty to 1st year med students in my class at Harvard. It is eye opening. Our schools benefit from the status quo, so we need outside pressure & new leadership to change things
comment in response to
post
This was in the works but the NIH announcement on Friday meant that they have gone into cost cutting overdrive starting today.
It’s tragic, these are good people; and it’s only just the beginning.
comment in response to
post
Shout out to all my coauthors and especially to Noah Jones, Mign-Chieh Shih and David Sontag! Always a privilege to work with smart and driven individuals
(7/7)
comment in response to
post
Finally, we performed a sensitivity analysis using an automated feature extraction package (clinicalml.github.io/omop-learn/) vs using domain-expert knowledge for model specification.
Results were similar, which is cool bc it lowers the barriers to performing these analyses more generally
(6/7)
comment in response to
post
Adverse event (AE) rates were the same, if not slightly better, for first-line antibiotics vs both second-line & alternative treatments. The one exception is a slight increase in skin-related AEs (ie rash); perhaps related to sulfa allergy?
(5/7)
comment in response to
post
First line treatments performed well vs second line treatments and better than alternatives (code for beta-lactams) with respect to primary outcomes (ie revisit within a month)
(4/7)
comment in response to
post
Here's our analytic method. We sought to adjust for differences that impact the likelihood of treatment and the likelihood of being lost to follow up using models optimized with ML algorithms
(3/7)
comment in response to
post
Here's our study design. We defined time 0 as the time an individual received treatment for UTI
(2/7)