I have a strong recollection of a study showing that #medicalAI studies (or devices) typically didn't report demographics of their data, and that the ones that did were biased. But I can't find it. Anyone know what I'm talking about?
Latest posts 5
AI is touted as a gamechanger for drug development across the whole cycle. Is it?
@LCDruedahl @AmeetSarpatwari, @TiMinCeBIL, & I find...maybe?
Modest use, mostly in earlier discovery stages. Lots of potential left!
@JAMANetworkOpen jamanetwork.com/journals/jaman…
#AI #drugs
@LCDruedahl @AmeetSarpatwari, @TiMinCeBIL, & I find...maybe?
Modest use, mostly in earlier discovery stages. Lots of potential left!
@JAMANetworkOpen jamanetwork.com/journals/jaman…
#AI #drugs
Khiara Bridges of @BerkeleyLaw has a great new piece laying out bias in medical AI used in maternal care—& making the surprisingly radical suggestion that it should be addressed through wide-ranging informed consent.
My @ireadjotwell review: cyber.jotwell.com/can-informed-c…
#medicalAI
cyber.jotwell.com Can Informed Consent Solve AI Bias? - Technology Law Khiara M. Bridges, Race in the Machine: Racial Disparities in Health and Medical AI, 110 Va. L. Rev. 243 (2024).Nicholson PriceArtificial intelligence (AI) is moving increasingly rapidly into health... My @ireadjotwell review: cyber.jotwell.com/can-informed-c…
#medicalAI
When a patient gets injured, who SHOULD be liable? (not who *is*, that's another thing).
@cohenprof and I have thoughts!
(spoiler: it's hospitals, with a huge caveat!)
Final version in DePaul Law Review out now:
ssrn.com/abstract=45177…
#ai #liability #medicalai
@cohenprof and I have thoughts!
(spoiler: it's hospitals, with a huge caveat!)
Final version in DePaul Law Review out now:
ssrn.com/abstract=45177…
#ai #liability #medicalai
How can we enable local health systems to collaboratively & effectively govern #medicalAI?
@kdpsinghlab, S. Bali, M. Sendak & I argue (today in
@NatMachIntell) that a) it's tough; b) some tasks will be unavoidably local; & c) policy can help.
Enjoy! @ rdcu.be/diYpU
@kdpsinghlab, S. Bali, M. Sendak & I argue (today in
@NatMachIntell) that a) it's tough; b) some tasks will be unavoidably local; & c) policy can help.
Enjoy! @ rdcu.be/diYpU