@Raamana_ 4 months ago
The Pitt Brainhack @DeBurghing starts with a Guest Lecture by @amarquand on "Using #machinelearning and big data neuroscience to move toward precision medicine in #psychiatry". Their lab does very interesting and important work - join us! Latest program: https://pitt-brainhack.github.io/DeBurghing2020/ https://t.co/jqZ6rC33i4
@raamana_ 1 year ago
tips to review #MachineLearning papers, esp. in #medicine & #neuroscience: -ask for a flowchart from raw data to final metrics! -don't assume CV (like repeated holdout) and advanced metrics (AUC) are implemented correctly. If you don't follow it drunk, they're likely wrong!
@raamana_ 1 year ago
Hi @geoffreyhinton, claiming #DeepLearning will replace radiologists in 5yrs is outrageous. I'll be happy to explain the challenges and why we the domain experts call it computer-AIDED diagnosis, not automated Dx. https://youtu.be/2HMPRXstSvQ #medicine #AI @GaryMarcus #MachineLearning
@raamana_ 2 years ago
Random tip: if a PI publishes over 25 papers/year (more than 2 papers/month), in a heavily data- and optimization-driven research e.g. #MachineLearning in medicine, without releasing their codebase or using public libraries, be HIGHLY skeptical about their results. #phdchat