Elias Eythorsson
@eliaseythorsson.bsky.social
Hospitalist. PhD in Epidemiology. Amateur statistician & prognostic modeler. Hope to grow up to be a trialist.
Reposted by Elias Eythorsson
Did you know? 2026 is one of the integers which, when divided by the sum of the squares of its digits, does nothing particularly remarkable
December 31, 2025 at 5:28 AM
Did you know? 2026 is one of the integers which, when divided by the sum of the squares of its digits, does nothing particularly remarkable
Reposted by Elias Eythorsson
We're doing the final sprint, and I think we're able to send the PDF of the forthcoming Bayesian Workflow book to the publisher in the next two weeks (500+ pages), which would mean it would be published some time next year
December 16, 2025 at 5:21 PM
We're doing the final sprint, and I think we're able to send the PDF of the forthcoming Bayesian Workflow book to the publisher in the next two weeks (500+ pages), which would mean it would be published some time next year
Reposted by Elias Eythorsson
Doing a bottom-to-top close edit now for errors and clarity. Just about 100 pages to go, not quite yet insane. Like some wetware LLM, I have been trained deeply on the distinctive text stylings of my individual coauthors. Feel I could produce novel utterances in any of their voices. Almost there!
We're doing the final sprint, and I think we're able to send the PDF of the forthcoming Bayesian Workflow book to the publisher in the next two weeks (500+ pages), which would mean it would be published some time next year
December 29, 2025 at 1:05 PM
Doing a bottom-to-top close edit now for errors and clarity. Just about 100 pages to go, not quite yet insane. Like some wetware LLM, I have been trained deeply on the distinctive text stylings of my individual coauthors. Feel I could produce novel utterances in any of their voices. Almost there!
Reposted by Elias Eythorsson
I am just learning of this 2015 retraction, adding to my "science as amateur software engineering" files. Seems they classified missing values as obs outcome of interest (divorce). Classified 32% of sample divorced, rather than true 5%. retractionwatch.com/2015/07/21/t...
December 18, 2025 at 9:06 AM
I am just learning of this 2015 retraction, adding to my "science as amateur software engineering" files. Seems they classified missing values as obs outcome of interest (divorce). Classified 32% of sample divorced, rather than true 5%. retractionwatch.com/2015/07/21/t...
Reposted by Elias Eythorsson
Ever stared at a table of regression coefficients & wondered what you're doing with your life?
Very excited to share this gentle introduction to another way of making sense of statistical models (w @vincentab.bsky.social)
Preprint: doi.org/10.31234/osf...
Website: j-rohrer.github.io/marginal-psy...
Very excited to share this gentle introduction to another way of making sense of statistical models (w @vincentab.bsky.social)
Preprint: doi.org/10.31234/osf...
Website: j-rohrer.github.io/marginal-psy...
August 25, 2025 at 11:49 AM
Ever stared at a table of regression coefficients & wondered what you're doing with your life?
Very excited to share this gentle introduction to another way of making sense of statistical models (w @vincentab.bsky.social)
Preprint: doi.org/10.31234/osf...
Website: j-rohrer.github.io/marginal-psy...
Very excited to share this gentle introduction to another way of making sense of statistical models (w @vincentab.bsky.social)
Preprint: doi.org/10.31234/osf...
Website: j-rohrer.github.io/marginal-psy...
Reposted by Elias Eythorsson
Reposted by Elias Eythorsson
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...
- 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...
Evaluation of performance measures in predictive artificial intelligence models to support medical decisions: overview and guidance
Numerous measures have been proposed to illustrate the performance of predictive artificial
intelligence (AI) models. Selecting appropriate performance measures is essential
for predictive AI models i...
www.thelancet.com
December 13, 2025 at 2:04 PM
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...
- 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...
Reposted by Elias Eythorsson
My love for em dashes is as old as the hills and as mighty as my inability to make a transition between thoughts literally any other way and I will be damned if I let the clankers take that from me
December 10, 2025 at 3:00 AM
My love for em dashes is as old as the hills and as mighty as my inability to make a transition between thoughts literally any other way and I will be damned if I let the clankers take that from me
Reposted by Elias Eythorsson
thank you yes that answers my question
December 5, 2025 at 4:08 AM
thank you yes that answers my question
Reposted by Elias Eythorsson
Reposted by Elias Eythorsson
If you have a manuscript ready to submit in mid-December do you:
a) Submit in December (when editors are bombarded with manuscripts)
b) Hold off until mid-January to let the editors have a damn holiday already
a) Submit in December (when editors are bombarded with manuscripts)
b) Hold off until mid-January to let the editors have a damn holiday already
November 28, 2025 at 11:30 AM
If you have a manuscript ready to submit in mid-December do you:
a) Submit in December (when editors are bombarded with manuscripts)
b) Hold off until mid-January to let the editors have a damn holiday already
a) Submit in December (when editors are bombarded with manuscripts)
b) Hold off until mid-January to let the editors have a damn holiday already
Reposted by Elias Eythorsson
{DAGassist} let's you process a DAG and estimate models giving {dagitty} input and baseline model #CausalSky #rstats
cran.r-project.org/web/packages... I think {ggdag} has more to offer for analysis of a DAG, while {DAGassist} is a one-stop package for classifying variables and getting estimates
cran.r-project.org/web/packages... I think {ggdag} has more to offer for analysis of a DAG, while {DAGassist} is a one-stop package for classifying variables and getting estimates
October 30, 2025 at 7:16 PM
{DAGassist} let's you process a DAG and estimate models giving {dagitty} input and baseline model #CausalSky #rstats
cran.r-project.org/web/packages... I think {ggdag} has more to offer for analysis of a DAG, while {DAGassist} is a one-stop package for classifying variables and getting estimates
cran.r-project.org/web/packages... I think {ggdag} has more to offer for analysis of a DAG, while {DAGassist} is a one-stop package for classifying variables and getting estimates
Reposted by Elias Eythorsson
every single tech idea is like “soon our robots will be capable of playing catch with your kid, freeing you up to spend more time working on your employers’ spreadsheets”
October 11, 2025 at 2:07 PM
every single tech idea is like “soon our robots will be capable of playing catch with your kid, freeing you up to spend more time working on your employers’ spreadsheets”
Reposted by Elias Eythorsson
you know what fair
October 10, 2025 at 9:35 AM
you know what fair
Reposted by Elias Eythorsson
CHATGPT: I understand where you're coming from. You worked really hard to get here, and now it's time to enjoy the fruit of your labors.
ISILDUR: So I should keep it? Elrond says I shouldn't
CHATGPT: The ring is precious. Sometimes friends don't have your best interests at heart.
ISILDUR: true
ISILDUR: So I should keep it? Elrond says I shouldn't
CHATGPT: The ring is precious. Sometimes friends don't have your best interests at heart.
ISILDUR: true
October 6, 2025 at 1:58 PM
CHATGPT: I understand where you're coming from. You worked really hard to get here, and now it's time to enjoy the fruit of your labors.
ISILDUR: So I should keep it? Elrond says I shouldn't
CHATGPT: The ring is precious. Sometimes friends don't have your best interests at heart.
ISILDUR: true
ISILDUR: So I should keep it? Elrond says I shouldn't
CHATGPT: The ring is precious. Sometimes friends don't have your best interests at heart.
ISILDUR: true
Reposted by Elias Eythorsson
We didn't randomize, and there was no allocation concealment or blinding, and we can't really be sure what intervention they got or how the outcomes were measured, but we emulated a trial by drawing a DAG.
August 31, 2025 at 3:21 PM
We didn't randomize, and there was no allocation concealment or blinding, and we can't really be sure what intervention they got or how the outcomes were measured, but we emulated a trial by drawing a DAG.
Reposted by Elias Eythorsson
Perfectly stated Darren. Don’t anyone think this is an exaggeration. I’ve seen this in supposedly reputable cardiology journals, sometimes even with omission of “by drawing a DAG”.
We didn't randomize, and there was no allocation concealment or blinding, and we can't really be sure what intervention they got or how the outcomes were measured, but we emulated a trial by drawing a DAG.
September 1, 2025 at 12:57 PM
Perfectly stated Darren. Don’t anyone think this is an exaggeration. I’ve seen this in supposedly reputable cardiology journals, sometimes even with omission of “by drawing a DAG”.
Reposted by Elias Eythorsson
So, the Cuban studies reporting near complete block of #SARSCOV2 / #COVID19 infection after high-risk exposures with 1M units may have been valid after all.
40k units were used once a day and cut infections by 40% 👇
academic.oup.com/cid/advance-...
The Nazoferon studies are also nice
40k units were used once a day and cut infections by 40% 👇
academic.oup.com/cid/advance-...
The Nazoferon studies are also nice
August 30, 2025 at 3:42 PM
So, the Cuban studies reporting near complete block of #SARSCOV2 / #COVID19 infection after high-risk exposures with 1M units may have been valid after all.
40k units were used once a day and cut infections by 40% 👇
academic.oup.com/cid/advance-...
The Nazoferon studies are also nice
40k units were used once a day and cut infections by 40% 👇
academic.oup.com/cid/advance-...
The Nazoferon studies are also nice
Reposted by Elias Eythorsson
Decongestion Secrets ("ammo" for rounds):
In a multi-center study of 283 acute heart failure patients, changes in renal filtration markers (cystatin C or creatinine) with aggressive diuresis were not associated with changes in markers of renal tubular injury (NAG, NGAL, or KIM-1)
In a multi-center study of 283 acute heart failure patients, changes in renal filtration markers (cystatin C or creatinine) with aggressive diuresis were not associated with changes in markers of renal tubular injury (NAG, NGAL, or KIM-1)
August 27, 2025 at 12:53 PM
Decongestion Secrets ("ammo" for rounds):
In a multi-center study of 283 acute heart failure patients, changes in renal filtration markers (cystatin C or creatinine) with aggressive diuresis were not associated with changes in markers of renal tubular injury (NAG, NGAL, or KIM-1)
In a multi-center study of 283 acute heart failure patients, changes in renal filtration markers (cystatin C or creatinine) with aggressive diuresis were not associated with changes in markers of renal tubular injury (NAG, NGAL, or KIM-1)
Reposted by Elias Eythorsson
The paper about splines in R that I wish I had known about when I was learning splines in R.
bmcmedresmethodol.biomedcentral.com/articles/10....
bmcmedresmethodol.biomedcentral.com/articles/10....
A review of spline function procedures in R - BMC Medical Research Methodology
Background With progress on both the theoretical and the computational fronts the use of spline modelling has become an established tool in statistical regression analysis. An important issue in splin...
bmcmedresmethodol.biomedcentral.com
August 22, 2025 at 4:51 PM
The paper about splines in R that I wish I had known about when I was learning splines in R.
bmcmedresmethodol.biomedcentral.com/articles/10....
bmcmedresmethodol.biomedcentral.com/articles/10....
Reposted by Elias Eythorsson
ICU #POCUS "Secrets":
This is why some times we achieve perfect alignment of the cardiac probe with the interventricular septum (IVS) & record great 4-chamber clips and other times* we cannot even find the IVS (when there is AIR interposed between the heart apex and the probe):
This is why some times we achieve perfect alignment of the cardiac probe with the interventricular septum (IVS) & record great 4-chamber clips and other times* we cannot even find the IVS (when there is AIR interposed between the heart apex and the probe):
August 21, 2025 at 5:30 PM
ICU #POCUS "Secrets":
This is why some times we achieve perfect alignment of the cardiac probe with the interventricular septum (IVS) & record great 4-chamber clips and other times* we cannot even find the IVS (when there is AIR interposed between the heart apex and the probe):
This is why some times we achieve perfect alignment of the cardiac probe with the interventricular septum (IVS) & record great 4-chamber clips and other times* we cannot even find the IVS (when there is AIR interposed between the heart apex and the probe):
Reposted by Elias Eythorsson
Now on CRAN, ggdiagram is a #ggplot2 extension that draws diagrams programmatically in #Rstats. Allows for precise control in how objects, labels, and equations are placed in relation to each other.
wjschne.github.io/ggdiagram/ar...
wjschne.github.io/ggdiagram/ar...
August 20, 2025 at 10:43 AM
Now on CRAN, ggdiagram is a #ggplot2 extension that draws diagrams programmatically in #Rstats. Allows for precise control in how objects, labels, and equations are placed in relation to each other.
wjschne.github.io/ggdiagram/ar...
wjschne.github.io/ggdiagram/ar...
Reposted by Elias Eythorsson
Reposted by Elias Eythorsson
New paper in @annalsofim.bsky.social
"50 ways to misinterpret clinical prediction models for treatment decisions”
--> Published version: www.acpjournals.org/doi/10.7326/...
--> Open access version: arxiv.org/pdf/2402.17366
"50 ways to misinterpret clinical prediction models for treatment decisions”
--> Published version: www.acpjournals.org/doi/10.7326/...
--> Open access version: arxiv.org/pdf/2402.17366
The Risks of Risk Assessment: Causal Blind Spots When Using Prediction Models for Treatment Decisions | Annals of Internal Medicine
Clinicians increasingly rely on prediction models to guide treatment choices. Most prediction models, however, are developed using observational data that include some patients who have already receiv...
www.acpjournals.org
August 11, 2025 at 2:32 PM
New paper in @annalsofim.bsky.social
"50 ways to misinterpret clinical prediction models for treatment decisions”
--> Published version: www.acpjournals.org/doi/10.7326/...
--> Open access version: arxiv.org/pdf/2402.17366
"50 ways to misinterpret clinical prediction models for treatment decisions”
--> Published version: www.acpjournals.org/doi/10.7326/...
--> Open access version: arxiv.org/pdf/2402.17366
Reposted by Elias Eythorsson
CVICU attending surveying the undifferentiated chaos of the MICU
August 4, 2025 at 1:37 PM
CVICU attending surveying the undifferentiated chaos of the MICU