A friend drove two hours today for her breast cancer chemotherapy only to learn it was canceled due to NIH cuts. Clinical trial ended midstream. No treatment, no restart, no plan, no information. Her bro-in-law's chemo was also canceled.
A friend drove two hours today for her breast cancer chemotherapy only to learn it was canceled due to NIH cuts. Clinical trial ended midstream. No treatment, no restart, no plan, no information. Her bro-in-law's chemo was also canceled.
Some examples of what VoxelPrompt can do!
Led by Andrew Hoopes (ahoopes.github.io), also with Victor Butoi, John Guttag and myself.
At @csail.mit.edu and the The MGH/HST Martinos Center for Biomedical Imaging
This flexible framework excels across many neuroimaging tasks, enabling a wide range of clinical and research goals.
For a first demonstration, we focus on the large scope of neuroimage analysis. VoxelPrompt can characterize and segment hundreds of anatomical and disease regions-of-interest, support various morphological measurements (e.g. tumor growth over visits), and produce language-based characteristics.
A core technical contribution is jointly learning the language-based agent with the adaptive vision network, which leads to a powerful mechanism for flexibly addressing complex analytical targets.
VoxelPrompt uses a language agent that interprets user prompts, predicts step-by-step instructions (as code), and employs a powerful vision network to analyze any number of patient-specific scans at once.
Most practitioners lack compute resources or machine-learning expertise to build new models for their specific needs. This creates significant barriers to translating AI advancements into scientific and clinical workflows.
Clinicians and scientists face a broad range of analytical imaging aims, yet most AI medical imaging tools used in practice are designed for narrow, predefined computational goals and data modalities.
Introducing VoxelPrompt, an agent-driven model that can tackle a landscape of diverse radiological tasks, extending far beyond those addressed by todayβs specialized models. VoxelPrompt addresses complex objectives with a solution grounded in interpretable vision operations.
Paper: lnkd.in/ezWSX6zQ
Recently thought about a model which takes a delineation instruction / segmentation protocol as input and provides the respective segmentation(s). Today I read VoxelPrompt. Super interesting work by Andrew Hoopes, Victor Ion Butoi, John Guttag and @adriandalca.bsky.social.
Weβve tried to shelter our minds but tonight weβre preparing for the genocide of Science.
Entire NIH departments will have no leaders.
World renowned Scientists kicked to the curb without cause, & grants will loose their lead investigators.
Is this what they want?
Chaos? Destruction?
Inside the collapse at the NIH
www.theatlantic.com/health/archi...
Meanwhile, in Canada, Pierre Poilievre is on record opposing vaccine mandates and public health measures in future epidemics/pandemics, and openly supported antivaxxers who terrorized Ottawa and shut down borders.
Pierre Poilievre is simply a danger to the health of Canadians.
Inexplicably, US administration has launched an unprecedented assault on science, on research institutions & on vital international organizations & initiatives.
An assault on science anywhere is an assault on science everywhere; the global community must unite
π§ͺ
www.nature.com/articles/d41...
Let me be clear. DOGEβs chokehold on the NIH will impact every single person. Holding back biomedical research this way, benefits nobody.
Unfortunately it won't be felt as a sharp economic shock where people can easily identify the cause and course-correct. It will be a slow stagnation until one day you realize you've been eclipsed but can't figure out why.
Americans are not remotely ready for the kind of economic downturn we face from kneecapping our research powerhouses. The scariest part is how few see what is coming and how it might take a decade or more to correct.
Let me be very clear. US academia has led the world because (a) it can attract international students and researchers and (b) has the resources to support long-term science. Both of these pillars are under severe threat with the current administration in Washington.
For @technologyreview.com I spoke with more than 10+ federal workers who occupy, or until recently occupied, scientific and technical positions across various agencies. They explained to me how much behind-the-scenes work the US government does to keep Americaβs engine of innovation humming. 2/
For decades, the US government has painstakingly kept American science #1 globallyβand every facet of American life has improved because of it. The internet? Flu shot? Ozempic? All grew out of federally-funded research. Now all that's being dismantled. 1/ www.technologyreview.com/2025/02/21/1...
Once upon a time I wrote a somewhat viral twitter thread on how to keep up with the rapidly growing scientific literature. I was wrong. Itβs gone too far. Stop obsessing over the latest arxiv drop. Cultivate your own curiosity and creativity.
So.
It has officially happened. I have been officially let go from my position at CDC. I wasnβt a probationary employee. I had stellar performance reviews every year.
Iβll be ok. But please understand that many worked HARD to get into these positions and were doing great public health work.
Weill Cornell Radiology is now accepting applications for the Cornell NYC AIMI Fellowship! A prestigious 3-year research position for recent Ph.D. grads in AI for medical imaging.
π§΅π
Welcome @melbajournal.bsky.social !
100 papers published to-date, and counting! π
@adriandalca.bsky.social @msabuncu.bsky.social
During one of the most severe flu seasons in recent years, too.
We have already had children die of influenza this year.
Influenza can be severe, and the average severity varies from year to year.
Being vaccinated would have been protective against the worst outcomes.
The world's engine of biomedical research is grinding to a halt π’Vast majority of modern vaccines, therapeutics, and diagnostics can trace their origin to NIH-funded research. This will have dire long-term effects.
U.S. Dept of Veterans Affairs confirms it has dismissed more than 1,000 employees
A once unthinkable concept for an agency that operates 100+ major hospitals and many hundreds more clinics
And processes veterans benefits
news.va.gov/press-room/v...
Putting RFK Jr. in charge of the nationβs public health is a huge mistake.
When dangerous diseases resurface and people canβt access lifesaving vaccines, all Americans will suffer β and RFK Jr.βs family could keep getting richer thanks to his serious conflicts of interest.
From the NYT article: "One person familiar with the decision said Dr. Tabak had been confronted with a reassignment that he viewed as unacceptable."