Randy Ellis
@randalljellis
Postdoc @chiragjp @HarvardDBMI | PhD @HurdLab @SinaiBrain. Give me anonymous feedback: https://t.co/g3M5Nbx0P5
ID:27503534
http://randalljellis.github.io 29-03-2009 21:04:29
3,4K Tweets
1,1K Followers
1,7K Following
'Why Scientific Fraud Is Suddenly Everywhere:' Our Ivan Oransky talks to Kevin T. Dugan of New York Magazine, in a conversation referencing Adam Marcus Carl Elliott Ben Willem Mol and more. nymag.com/intelligencer/…
Can we reconstruct images from brain activity? The Yuki Kamitani lab argues that recent work uses a non-diverse dataset, and with appropriate controls, performance breaks down completely. This has important implications for what we can learn from such datasets.
arxiv.org/abs/2405.10078
We recently published a multivariable prediction model to aid in the decision to obtain a bone marrow sample in MGUS acpjournals.org/doi/10.7326/M2… . Now I seek critique of my paper & methods by experts in prediction modelling on Frank Harrell's datamethods
discourse.datamethods.org/t/critique-of-…
A registered report published in Science Advances following recommendation by PCI Registered Reports (@[email protected]) science.org/doi/10.1126/sc…
Hard - if not impossible - to find examples of bespoke models (cc Richard McElreath 🦔) in medical research
But here is an example by Keith Goldfeld et al
Amazing work on using simulation + prior predictive check + bias analysis to adjust priors choice!!
pubmed.ncbi.nlm.nih.gov/36698073/
Thanks to Society for Neuroscience (SfN) for the chance to participate in this webinar on #registeredreports and replicability, and my co-panelists Christophe Bernard (@eNeuroEiC) and Jose Abisambra (Co-EiC of Brain Research). Both eNeuro and Brain Research offer RRs! Watch: neuronline.sfn.org/scientific-res…
If grants/pubs remain the currency of science, #registeredreports are the solution. Publishing null results in repositories is great, but if journals won’t, it won’t help careers. RRs solve both problems and PCI Registered Reports (@[email protected]) are leading the way. nature.com/articles/d4158…
Starting in an Hour! ⏰
Discuss replicability issues in science and how #RegisteredReports can support rigorous & reproducible research with Randall Ellis, PhD, (Randy Ellis), Christophe Bernard, PhD, #eNeuro Editor-in-Chief (@eNeuroEiC), and Jose Abisambra, PhD,
Don't forget to register for this free webinar on #RegisteredReports supporting reproducible research
Registered reports can fix this by realigning academic incentives. To learn more about #registeredreports , register to hear our conversation about them hosted by Society for Neuroscience (SfN) on Wednesday at noon ET: neuronline.sfn.org/scientific-res…