On National Cancer Survivors Day we want to highlight our amazing Living Hope community. Living Hope is our community of cancer survivors and cancer patients who are dedicated to VeloSano and are always #1OOPercentForTheCure. Learn more about Living Hope: bit.ly/3z3ZWoS

TWEETORIAL linked here: x.com/CancerConnecto… #clecliniccancer Moffitt Research Cleveland Clinic Lerner Research Institute CWRU School of Medicine MSTP at Case Western Reserve University Cleveland Clinic MD

Jake Scott CWRU School of Medicine MSTP at Case Western Reserve University OncoAlert The Lancet Oncology Cleveland Clinic MD Congrats team!

Estimating patients' individualized probability of an event of interest is challenging. A must-read book by Gerds and mk clearly explains and guides you when developing and validating prediction models. #predictionmodels #machinelearning

Thanks for comments in The Scientist article Melvin (Ane Appelt & Wendy Woodward MDPhD & Alejandra Manjarrez). One thing that's confusing though: the HR for GARD (close to 1, as you point out, 0.98) is PER UNIT GARD -- this is a continuous analysis. cont. thelancet.com/journals/lanon…

while each patient requires a different amount of physical dose to increase 1 unit of GARD (based on radiosensitivity), imagine a scenario where you dose-escalate a patient by 10 GARD, this would yield a predicted HR = 0.98^10 = 0.82 and so on for other GARD changes

the point of the nomogram in figure 4 is to show this, but we put things back into the ABSOLUTE world of each disease (rather than the abstraction needed for the pooled analysis where everything was RELATIVE).

So many "regression versus machine learning" comparisons are not fair, as they don't allow non-linear trends for predictors in the regression model (eg using splines or polynomials). As such, they only serve to reveal a lack of understanding of good statistical practice

Thank you Society for Medical Decision Making (SMDM) for this amazing honor! #SMDM21

Fantastic work by Dennis Lal and team! lerner.ccf.org/news/details/?… #epilepsy #genetics

Most analyses that found response a surrogate ignore immortal time bias. Congratulations @sujataCLE and Julio Garcia Aguilar for a rigorous analysis that avoids this common trap.

NEW PREPRINT Assessing performance and clinical usefulness in prediction models with survival outcomes: practical guidance for Cox proportional hazards models (with SAS and R code) medrxiv.org/content/10.110…

We can already anticipate the (aging) objections, but the idea of re-branding at least a subset of GS6 seems to be picking up momentum. Feels like the right side of history :) Scott Eggener⁩ ⁦Alejandro (Ale) Berlin⁩ ⁦Andrew Vickers ⁦Howard Wolinsky⁩ ascopubs.org/doi/full/10.12…

Computable phenotype of disease health states transition probabilities b/w health states estimates real-world impact! Jacob Kurowski Jean-Paul Achkar Cleveland Clinic Lerner Research Institute bit.ly/3HLDUf2

give.velosano.org/fundraiser/396…

This week on the Cleveland Clinic Cancer Advances podcast I talk to Cleveland Clinic MD radiation oncologist Rahul Tendulkar (Rahul Tendulkar, MD) and Chair Dept Quantitative Health Sciences mk about use of nomograms for patients with #cancer my.clevelandclinic.org/podcasts/cance…

Great to see old friends!

The Society for Medical Decision Making (SMDM) Career Achievement Award winner is Michael Kattan Cleveland Clinic Lerner Research Institute #SMDM22 (4/4) Thank you to everyone who nominated members for our awards and congratulations to all of this year's winners! 🏆

Thanks for the clarification. I forgot our url changed

Nothing prettier to us than a full Bike to Cure starting line ready to take off! 😍 Have you registered yet? bit.ly/3gCfWq8