PhilosopherMD (@philosophermd1) 's Twitter Profile
PhilosopherMD

@philosophermd1

This is my no politics Twitter account. I need some sanity occasionally. Physician & Meta-Physician #Theology #Philosophy #Oncology #Statistics #Investing

ID: 1103158619284090880

calendar_today06-03-2019 05:01:25

1,1K Tweet

408 Followers

260 Following

PhilosopherMD (@philosophermd1) 's Twitter Profile Photo

Does anybody know the gr4 rate for prostate? 0415 only 1 gr4 and PACE-B had 0. 10 patients with Gr4 w/protons. A little less gr 1-3 toxicity but a higher risk of Gr4 toxicity - Doesn’t seem worth it. The paper below states Gr4 should be a “zero event” but still calls H+ safe 🤔

Susannah Ellsworth (@free_radical28) 's Twitter Profile Photo

"Black swan events" (eg change in supervision req's) & decline in medicare population also likely to depress demand side. Per excellent discussion by Chirag Shah residency contraction may be needed to better match supply & demand

Frank Harrell (@f2harrell) 's Twitter Profile Photo

Fantastic new paper on sample size requirements of statistical models vs. #MachineLearning , except for not giving proper emphasis to smooth calibration curves: onlinelibrary.wiley.com/doi/10.1002/si… #Statistics

PhilosopherMD (@philosophermd1) 's Twitter Profile Photo

Crazy how Twitter lets us look at these back and forth arguments from these renowned experts! I highly recommend going through these threads - guaranteed to make you smarter 🧠 Stephen John Senn Frank Harrell Judea Pearl

Matt Spraker (@sprakermdphd) 's Twitter Profile Photo

PhilosopherMD Nadeem Riaz C. Jillian Tsai, MD, PhD Nancy Lee I know a lot of people that thought it was kinda nuts it became category 1 NCCN off a tiny single arm study. Pointing out that things were done poorly in another subsection of oncology is not a good argument for doing another thing poorly.

Amit Chowdhry, MD, PhD (@amit_chowdhry) 's Twitter Profile Photo

Our review on Multiplicity in Clinical Trials is now out in the Red Journal, with NRG Oncology Stephanie Pugh, John Park VA Kansas City, John Kang Kuan G. S. #radonc #stats redjournal.org/article/S0360-…

Prof. Nikolai Slavov (@slavov_n) 's Twitter Profile Photo

Biology has more data than ever before. We need to focus on interpreting these data. On the logical connections & the supportable conclusions. Falsifiable models can lead to reliable models.

Biology has more data than ever before.

We need to focus on interpreting these data.

On the logical connections & the supportable conclusions.

Falsifiable models can lead to reliable models.
PhilosopherMD (@philosophermd1) 's Twitter Profile Photo

Genuine question, why not run these lympho XRT trials just a single arm trial and say if PFS is >95% then we will omit XRT. Something in me says it’s unethical to randomize and disregard your stat endpts, if that’s the case why waste so many patients time? ascopubs.org/doi/10.1200/JC…

PhilosopherMD (@philosophermd1) 's Twitter Profile Photo

Although I’m no Bayesian this is a great and challenging post by Frank Harrell. I’ve learned a lot by going down the Bayesian rabbit hole and am better for it.