Denis Mongin
banner
denismongin.bsky.social
Denis Mongin
@denismongin.bsky.social
Started with lasers and nonlinear optics, now with biomedical informatics, data and statistics in health sciences.
Research in #Metaresearch, #rheumatology and #publichealth at HUG and University of Geneva.
#Rstats ninja, #Openscience
Pinned
Do people register their #randomizedtrials before starting it ?
We studied more than 1000 RCT in #rheumatology to answer this question, to see how it evolved in time and how it varied between RCTs⬇️

Just published in @jclinepi.bsky.social :
jclinepi.com/article/S089...

#Metaresearch #EpiSky
Prospective registration of trials: where we are, why, and how we could get better
Transparent trial conduct requires prospective registration of a randomized controlled clinical trial (RCT) before the enrollment of the first participant. We aimed to (1) estimate the proportion of R...
jclinepi.com
Dear science fellows
Has anyone tried sci-net ? (aka has paid the sci-hub token to have the invite)
November 27, 2025 at 9:36 AM
Vidéo de @gmilgram.bsky.social qui aborde un problème central et actuel de la médecine universitaire: la contamination par les pratiques ésotériques.
Ici un niveau D.E.L.I.R.A.N.T

www.youtube.com/watch?v=ILTg...
Quand le bullshit s'infiltre à l'hôpital 😤 (Enquête)
YouTube video by G Milgram
www.youtube.com
November 27, 2025 at 8:01 AM
@mathishammel.bsky.social je me souviens d'avoir vu un post de vous parlant de comment éviter les blocage pour du scraping massif.
Est-ce qu'on peut se parler en DM ? J'ai un projet de recherche, et je bloque sur un détail technique de scraping avec Selenium
November 26, 2025 at 12:38 PM
Reposted by Denis Mongin
DOGE did not fail in any way to accomplish its goals.

Its goals were never efficiency or saving money.

Its goals were to destroy as much of government as possible forever, and to steal data for the Space Nazi.

DOGE is fading away like bank robbery gangs fade away after the robberies are done.
November 23, 2025 at 12:26 PM
One of the wildest impact of climate change on human society recently (and we hear little about it): Tehran is not viable anymore because of the lack of Water. 10 millions people.

They are thinking about relocating Iran capital somewhere else

peterfrankopan.substack.com/p/makran-or-...
Makran or Bust: Tehran's water crisis gets worse
Ten days ago I wrote that Tehran was approaching a point where warnings, pressure cuts and appeals to save water would no longer be enough.
peterfrankopan.substack.com
November 23, 2025 at 2:21 PM
Reposted by Denis Mongin
J'ai ma réponse : le 19 novembre 2025...
November 21, 2025 at 7:39 PM
Impressive source of dataviz knowledge

@annalombardi.bsky.social , if you don't have it
November 19, 2025 at 1:45 PM
Just learned that chatgpt was down by visiting bluesky
😁😁
Still, spare a thought and or prayer for all the ChatGPT users currently unable to chat.
November 18, 2025 at 3:09 PM
Just published in Research Synthesis Methods

How to be more accurate than humans at abstract screening tasks using several small open-weight LLM?

doi.org/10.1017/rsm....

We actually circumvented the "LLM lie and hallucinate" problem for simple text analysis tasks !!

Let me explain

⬇️1/8
Beyond human gold standards: A multimodel framework for automated abstract classification and information extraction | Research Synthesis Methods | Cambridge Core
Beyond human gold standards: A multimodel framework for automated abstract classification and information extraction
doi.org
November 18, 2025 at 9:30 AM
Reposted by Denis Mongin
La Corse, bloc par bloc

On a replongé les mains dans les boites de Lego pour recréer la carte, mais en relief cette fois.

#30DayMapChallenge – J17 A New Tool

Retrouvez l'ensemble des créations des collègues : www.ign.fr/mag/30DayMap...

✍️ Maureen Chatel et Marie Stretti
November 17, 2025 at 10:08 AM
Reposted by Denis Mongin
Day 11 #30DayMapChallenge - Minimal

Bounding boxes of states around the world

#rstats #dataviz
November 14, 2025 at 10:21 PM
Reposted by Denis Mongin
{ggalign} is patchwork on steroids, and has some ability to add lines between plots: yunuuuu.github.io/ggalign-book/
ggalign: Bridging the Grammar of Graphics and Complex layout
yunuuuu.github.io
November 14, 2025 at 2:28 AM
Reposted by Denis Mongin
#30DayMapChallenge Day 13 - 10 minute map

Reachable areas within a 10-minute walk / 10-min isochrones of Yamanote Line (山手線) stations

And yes, the map took much longer than 10 mins to make, sorry not sorry
November 13, 2025 at 3:29 PM
Reposted by Denis Mongin
This is one of the most remarkable academic debacles I've ever seen.

A large RCT got published in BMJ. There are currently 44 Pubpeer comments, mostly about the data, including...well. Read for yourself.
pubpeer.com/publications...
PubPeer - Prevention of acute myocardial infarction induced heart fail...
There are comments on PubPeer for publication: Prevention of acute myocardial infarction induced heart failure by intracoronary infusion of mesenchymal stem cells: phase 3 randomised clinical trial (P...
pubpeer.com
November 11, 2025 at 1:19 AM
Reposted by Denis Mongin
#Day07 of #30DayMapChallenge on Accessibility

This map shows the time needed to cycle from Bains des Paquis (a popular fondue and sauna place in Geneva)

🆕 1st time use of the `osmr` package to define #isochrones.

Gallery: guillaume-noblet.com/30DayMapChal...

#dataviz #rstats #ggplot2 #gis #map
November 11, 2025 at 10:41 AM
Reposted by Denis Mongin
Oops. Ooooooooooooops.

I do hope that nobody has been given or denied a job/promotion based on their SpringerNature citation counts in the past 15 years.

arxiv.org/pdf/2511.01675

h/t @nathlarigaldie.bsky.social
November 7, 2025 at 2:02 PM
I have a revision of an article to do.
The editor list starts with A
....
and finishes with AE

Because there were not enough letters in the alphabet for all the changes.
It is gonna be a long day
November 10, 2025 at 10:26 AM
Reposted by Denis Mongin
If you want to read up on mediation analysis, we have a paper explaining the underlying causal inference issues: journals.sagepub.com/doi/10.1177/.... If you wonder about why I do pay attention to the p values: www.the100.ci/2018/02/15/t...
The uncanny mountain: p-values between .01 and .10 are still a problem
[Update: After this post had been published, Uli Schimmack and I had a quick chat and Uli was very surprised to learn that I hadn't read his 2012 Psychological Methods paper on the topic. He has now p...
www.the100.ci
November 9, 2025 at 7:29 AM
Reposted by Denis Mongin
#30DayMapChallenge
Day 7: Accessibility

Frequency of routes to the nearest park in Santiago de Compostela. 🗺️

#rstats #dataviz
November 7, 2025 at 3:25 PM
I think we will know in a year or two max, when the AI bubble will have exploded and US will have thousand of GPU worth nothing
A fun question. In 5 years time, what looks better? The US’s enormous bet & capex on AI? Or China’s equally enormous bet and capex on renewables?
China has made cheap, clean energy available in huge quantities. The world should take the win econ.st/4oqFszB

Photo: Eyevine
November 7, 2025 at 1:48 PM
Reposted by Denis Mongin
I think about this post every day 🧪
November 7, 2025 at 12:15 PM
Reposted by Denis Mongin
Day 05 of #30DayMapChallenge on Earth (Classical Elements 1/4)

I've discovered and used the new `tmap.mapgl` to show Swizerland's elevation. It's mindblowing how easy it is.

guillaume-noblet.com/30DayMapChal...

#dataviz #rstats #tmap #gis #mapgl cc @mtennekes and thanks!
November 7, 2025 at 12:30 PM
Review of one article submitted to jmir: in the infinite checklist, the editor explicitly asks to cite more jmir papers.
😶
Is this normal/common ?
November 7, 2025 at 9:24 AM
Reposted by Denis Mongin
A little bit of #ggplot2 and #rayshader for the "dimensions" prompt on day 6 of the #30DayMapChallenge! 🌧️

#RStats #DataViz
November 6, 2025 at 11:43 PM