Adam Sager πŸ‡ΊπŸ‡Έ (@Sager) header
Adam Sager πŸ‡ΊπŸ‡Έ (@Sager) profile picture

@OpenVet / saving animals from preventable death & misdiagnosis. The medical infra for every species. This window opens once in the history of medicine.

Miami, FL
Joined December 2009
Photo by Adam Sager πŸ‡ΊπŸ‡Έ (@Sager)Photo by Adam Sager πŸ‡ΊπŸ‡Έ (@Sager)Photo by Adam Sager πŸ‡ΊπŸ‡Έ (@Sager)Photo by Adam Sager πŸ‡ΊπŸ‡Έ (@Sager)Photo by Adam Sager πŸ‡ΊπŸ‡Έ (@Sager)Photo by Adam Sager πŸ‡ΊπŸ‡Έ (@Sager)Photo by Adam Sager πŸ‡ΊπŸ‡Έ (@Sager)Photo by Adam Sager πŸ‡ΊπŸ‡Έ (@Sager)Photo by Adam Sager πŸ‡ΊπŸ‡Έ (@Sager)Photo by Adam Sager πŸ‡ΊπŸ‡Έ (@Sager)
Adam Sager πŸ‡ΊπŸ‡Έ retweeted
π—’π—½π—²π—»π—©π—²π˜ π—₯π—Όπ˜‚π—»π—±π˜€, No. 11 Every week, veterinarians from thousands of animal hospitals and practices across the USA bring OpenVet their clinical questions. OpenVet's Chief Veterinary Officer, Dr. Natalie Marks, reviews the patterns and identifies what matters to practicing vets. π—§π—΅π—Άπ˜€ π˜„π—²π—²π—Έ: π‘™π‘’π‘π‘‘π‘œ 𝑖𝑠 π‘π‘Žπ‘π‘˜, π‘Žπ‘›π‘‘ 𝑖𝑑 β„Žπ‘Žπ‘  π‘Žπ‘› π‘Žπ‘‘π‘‘π‘Ÿπ‘’π‘ π‘ . Leptospirosis up 50%, from vets licensed in Iowa, Colorado, Tennessee and Michigan. Parvovirus up 44%, mostly Sun Belt. Heartworm and tick-borne questions fell. And more, including Natalie's read of the data. πŸ‘‡ Click below to read the weekly summary. The full report goes out every week to veterinarians who are OpenVet members. OpenVet membership is free for licensed veterinarians. Sign up: openvet.ai/?utm_source=round… @Sager @openvetai
4
4
23,001
Adam Sager πŸ‡ΊπŸ‡Έ retweeted
Three years ago today: 1,200 murdered. 251 taken hostage. The deadliest day for Jews since the Holocaust. Since then, antisemitism hasn't returned. It's exploded. Attacked in the streets. Hunted on campuses. Murdered from Washington to Sydney. For one reason: being Jewish. October 7th didn't create this hatred, it revealed it. I'm raising my daughters to be proud Jews. Loud. Visible. Unapologetic. We will not hide πŸ’™
129
310
28
3,570
94,579
Veterinary AI can recommend a treatment and never find out whether the animal got better. The prescription tells you what was planned. The next visit may tell you what happened. But what if there is no next visit? 1/9
πŸ€– Made with AI
1
65
Before a veterinary AI company says it learns from every case, it should be able to say how many cases have follow-up, what was measured, and which patients disappeared from view. Otherwise, it may be learning what gets prescribed without knowing what helps. 8/9
1
1
17
So the open questions vets have to answer: What follow-up information would make the biggest difference to your next treatment decision, and what could a busy clinic realistically collect? And are those right *questions* discoverable with data? That should help decide what veterinary AI builds next. 9/9
1
1
18
An AI scribe can influence a diagnosis without ever suggesting one. It happens earlier, when the conversation becomes the record. Whatever reads that record next inherits decisions about what was said, what mattered, and what could be left out. 1/8
πŸ€– Made with AI
1
1
2
46
That also creates an opportunity. A new result arrives. The system can revisit an earlier observation, check the original wording, and see whether the previous interpretation still holds. A summary that discarded the distinction gives it less to work with. 7/8
1
1
5
As scribes connect to diagnostics, their job expands. They are producing a record that other systems will reason over. The test is whether the next system can distinguish what was observed, what was inferred, and what nobody asked. 8/8
1
4
That also creates an opportunity. A new result arrives. The system can revisit an earlier observation, check the original wording, and see whether the previous interpretation still holds. A summary that discarded the distinction gives it less to work with. 7/8
1
9
As scribes connect to diagnostics, their job expands. They are producing a record that other systems will reason over. The test is whether the next system can distinguish what was observed, what was inferred, and what nobody asked. 8/8
1
1
14