Ultra processed foods are no Beuno
@aprilkhademi
Canada Research Chair in AI for Medical Imaging. Image Analysis in Medicine Lab @TorontoMet @UofTMedIm @VectorInst @UofT_TCAIREM @UnityHealthTO @iBESTResearch. https://www.torontomu.ca/akhademi/
Ultra processed foods are no Beuno
LOL -- wonder if Tylenol can sue him.
Monday mornings? Or Friday?
Times Higher Education (THE) Global AI Summit wrapped up yesterday and enjoyed taking part in the panel with esteemed colleagues: From Bench to Bedside: Revolutionising Healthcare using AI.
www.timeshighered-events.com/global-ai-su...
Hi Bluesky community. Me and my team are participating in a cancer hike fundraiser to bring awareness and raise funds for cancer research. Would love to have your support.
fundraisemyway.cancer.ca/participants...
Thank you for the shout out and support through the CCS grant programs! Itβs been an honour to work on AI tools that can improve breast cancer patient outcomes.
Interesting work. Authors developed an AI model to detect bilateral asymmetry as a breast cancer (BC) risk factor to predict BC risk up to 5yrs in advance. If both breasts had equal risk, how this would work? Either way, great tech & can't wait to see more like this.
www.rsna.org/news/2024/ma...
p-Tau217 is a breakthrough blood test that can predict and may help prevent Alzheimer's disease.
erictopol.substack.com/p/the-breakt...
Here's the rundown on it. Free, open-access.
Thank you for this great, informative article. This would be a much cheaper way to test for Alzheimerβs- with potential of finding it early too. Really great π
Thank you to the Canadian Cancer Society for funding my lab and trainees!
Congratulations, @aprilkhademi.bsky.social, on this great honor! ππ Weβre very proud to support your #research into using machine learning to improve #cancer diagnoses. We canβt wait to see the positive impact it will have!
#CancerResearch #AI
Two of our Ki67 algorithms were compared to commercial and open source solutions, using a prospective, fully-blinded (to us) dataset of 287 patients (2-3 samples per patient). Our Deep Learning based method, PiNET, was the most accurate and consistent on this diverse dataset.
A paper we worked on was awarded one of the five "Best of the USCAP - United States and Canadian Academy of Pathology Journals" in Laboratory Investigation for 2024.
www.laboratoryinvestigation.org/article/S002...
Thanks to my collaborators and IAMLAB team who made this work possible! We enjoyed taking part in this study, and demonstrating the robustness of our tools on prospective data.
@ccsresearch.bsky.social
Agreed. It is a bit slow to pick up.
The secret of life? Just enjoy the ride
π
This is some golden marketing. Right on brand.
βThe orange oneβ π
How is it possible to do this. lol. Arenβt you afraid of getting run over or running into something
Harvard scientist claims 'God is real': Reveals mathematical formula that could prove it
m.economictimes.com/magazines/pa...
π Turkey is the destination for hair transplants.
π
With a heavy heart, I've decided to suspend all academic travel to the USA for me and my lab. Given the escalation of tensions and uncertainties, it seems to be the wisest move. To our US colleagues, please be certain that we will continue to do what we can to support you. Science is global.
Interesting study looking at AI-based super resolution (SR) techniques for medical imaging. SR is low hanging fruit in the medical device space as it has a lower (regulatory) barrier to entry, and has a clear ROI (scanning more patients in less time).
www.emjreviews.com/radiology/ne...
Sticker given to me by a colleague. π