@bttyeo @twktan This is great! Brain age, as a concept, is quite nonsensical. It relies on assumed model behaviours (including out of distribution) and on the hypothetical alignment between age and the target of interest (eg disease). It’s fantastic to see its utility being debunked.

RT @KingsCollegeLon: King's scientists have used the Cambridge-1 supercomputer to create AI images of brains so accurate and realistic, tha…

Position 1: Real-world Healthcare Data - Position 2: Foundational Models for Healthcare - @ProjectMONAI, @AI4VBH

@ZiyunLiang_ This is ia very apt observation and very much true. Great paper! Z-score and KDE’s or integration over reconstructions been proposed as solutions to this. Have you looked into these?

@mertrory Didn’t know about this paper. Super cool writing.👌 giving some people time to think and reflect was probably one of the only good things about the pandemic.

RT @lena_maierhein: 🍻EXCITING DAY: Our @naturemethods sister publications on #MetricsReloaded are out🎉. Recommendations:

RT @jthteo: Cheater (AI shortcuts)

Fantastic work but Pearse and his team. These are exactly the kinds of models we need to move AI in healthcare forward. 🚀🚀

RT @pearsekeane: 1/ 🚨🚨 New paper alert 🚨🚨 Introducing RETFound, a foundation model for ophthalmology We’re super excited about this and…

Check out most recent work on Generative AI for medical imaging data, part of the new open source MONAI Generative Models package. #openscience

RT @Warvito: Good news in medical generative AI!🎉🎉 We just published the preprint of the MONAI Generative Models Extension! Check out our l…

RT @EceVanderbilt: The tutorial "Diffusion Models for Medical Imaging" at @MICCAI_Society will present an overview of generative modelling,…

@emrobSci The easiest thing is to pay full price and not get membership. Silly and inneficient, but the only safe way…

RT @sola_sesola2003: @EJR_official_ @EuSoMII @mad_rad_007 @ELS_Radiology @mjorgecardoso @aparpaleix @AditiRanjan_ Thanks @mad_rad_007 for…

RT @m__antonelli: Post-doctoral research associate position on Multimodal AI for Emotion Recognition at King's College London Closing dat…

@cwcyau @DrHughHarvey Payed to review?? Some academics would be rich!!

@DrHughHarvey It’s not about determinism; It’s about uncertainty and posteriors. Models that predict uncertainty bounds are often non-deterministic but provide mean/variance estimates of predictions. The problem is how to do this in language. Maybe proposing multiple sentences/completions?

RT @marksgraham_: Our work applying diffusion models to out-of-distribution detection has been accepted at the CVPR workshop on Visual Anom…

@sola_sesola2003 @mrm_highlights @ISMRM Congratulations Sola!!