Yogesh Rathi
@yogi_spoke
Associate Professor at Harvard; neuroscience; MRI, artificial intelligence; Yoga; alumni BITS Pilani, Georgia Tech
ID: 889876157935611904
http://pnl.bwh.harvard.edu/people/yogesh-rathi 25-07-2017 15:53:03
1,1K Tweet
988 Followers
350 Following
I guess only companies will be able to afford to apply machine learning and AI to the UK Biobank freezing out academic labs. Maybe that’s not the intent of this policy but that will be the consequence.
Super excited by this randomized double-blinded preregistered clinical trial: A network defined by treatment response from invasive stim (DBS) → used to noninvasively modulate the brain (tDCS). This improved motor functions in PD. Thread below 👇 Center For Brain Circuit Therapeutics Michael Fox
The future of AI is neurosymbolic. #longread about Google DeepMind’s new work (cc: Demis Hassabis) that is easily the most positive and important essay I have written this year. If you care about what happens in AI after the generative AI bubble bursts, please have a look.
My article in the The Times of India (The Times Of India) today, Sunday edition. "India needs reforms for higher education akin to the sweeping economic reforms of 1991. Three primary issues must be addressed to bring accountability to our educational institutions: faculty quality,
📢Preprint alert! ADC-#fMRI nicely complements BOLD in mapping motor pathway activation *and* recovers the functional organization of the visual dorsal & ventral streams. Congratulations Jasmine Nguyen-Duc et al! 🥳 #MicMapLab #FIREPATH 🧠🧲🫧 European Research Council (ERC) Faculté de biologie et de médecine (inactif) Swiss National Science Foundation
Will Consagra (UNC) estimates continuous connectivity (cf Daniel Moyer ‘s paper pubmed.ncbi.nlm.nih.gov/28487128/ ) but instead of using kernel density estimation, the product space function is learned with a deep neural net !
Mukund Rao The original take off for this work is the question on the nature of akasha, which is supposed to be the medium for consciousness. This is how I got there: medium.com/@subhashkak/ou…
Do you use multi-echo sequences for estimating the T2 in a diffusion-relaxation sequence ? Well - turns out there is bias in that estimation as we show in our new paper in MRM Highlights onlinelibrary.wiley.com/doi/10.1002/mr… — we also provide a neat solution — use of time-division
Excited to share our work on between-scanner test-retest reliability of ultra-low-field brain MRI morphometry and correspondence to high-field MRI: biorxiv.org/content/10.110… | We scanned 23 healthy adults (20-69 y) on two identical 64 mT Swoop® Portable MR Imaging® System–Hyperfine, Inc Swoop systems and 3 T MRI (1/n)
💡"Current status and future challenges in brain imaging: Insights from leading experts" 💡Prof Chao-Gan Yan Chao-Gan Yan interviewed 15 highly-cited brain mapping researchers to ask them about the current status, future directions in brain imaging sciencedirect.com/science/articl…