Jerzy Neyman (L) and Egon Pearson (R)
#OTD 1933 Egon Pearson & Jerzy Neyman publish 'On the problem of the most efficient tests of statistical hypotheses' introducing the Neyman-Pearson lemma: the most powerful test is the likelihood ratio test; the most efficient test controls error by maximizing power for a given significance level.
16.02.2026 10:47 — 👍 2 🔁 2 💬 0 📌 0
Our guidance regarding performance measures for medical AI models is finally out!
- Stop bashing AUROC, although it does not settle things
- Calibration and clinical utility are key
- Show risk distributions
- Classification statistics (e.g. F1) are improper
www.thelancet.com/journals/lan...
!!NEW PUBLICATION!!
www.ahajournals.org/doi/10.1161/...
!!NEW PUBLICATION!!
A total of 55 studies included (108,678 individuals) in meta-analysis. The pooled prevalence of dental fluorosis was estimated to be 34.5% (95%CI: 28.9–40.3) for all age groups, and in children (5-18 years) was 32.1% (95%CI: 26.6–37.9).
onlinelibrary.wiley.com/doi/10.1111/...
!!ANNOUNCEMENT!!
Prognosis Research In Healthcare Summer school 2025
lnkd.in/ehgZZtfF
The course will take place between the 9th and 11th of July 2025 and will be hosted online through a dedicated custom platform.
REGISTRATION DEADLINE: 30-06-2025
@erik-van-zwet.bsky.social has written a very nice blog discourse.datamethods.org/t/dichotomiz...
on the loss in dichotomising continuous outcomes. There is also a shiny app here vanzwet.shinyapps.io/info_loss/
"Don't cut your measures, cut the habit."
