iiinova — where your digital self keeps evolving

Raza @transfusion.health · 24/05/2026

I miss Mart Janssen. A brilliant data scientist and wonderful human. I still read his papers and miss him.

Raza @transfusion.health · 28/01/2026

Thank you for the resource. I wish biostatisticians would publish some sort of consensus guidelines/heuristics that us dumb clinicians can rely on to check that we are mostly doing the right thing (I am told statisticians will never agree enough to produce such a thing). An old paper I did like:

220 2

Raza @transfusion.health · 09/12/2025

The way the exponential form generalizes so many common, seemingly unrelated distributions (and all the properties you get for free, like sufficiency) has to be one of the most elegant things I’ve learned in #stats, and in general Good reader: brunomaga.github.io/Exponential-...

050 5

Raza @transfusion.health · 26/10/2025

For those wrestling with applying MINT trial and new AABB RBC MI guidelines results with nuance, I think this figure strikes a good balance for navigating the evidence base at the bedside doi.org/10.1016/j.cj... #transfusion #medsky #cardiology

010 1

Raza @transfusion.health · 22/09/2025

Classic teaching is that no post-1 hr increment points to immune refractoriness. One important exception is device-related failures, e.g. plts going into a clotted line, a faulty dialysis/apheresis circuit, an ECMO machine, etc! #hemesky #transfusion

040 4

Raza @transfusion.health · 14/06/2025

I was today years old when I finally saw this masterpiece, which is an embarrassing confession as a movie buff. Absolutely blown away by the writing and pacing #filmsky

030 3

Raza @transfusion.health · 03/05/2025

Can we rationalize the way we give diuretic prophylaxis before transfusion? Colleagues @uhntoronto.bsky.social incl talented co-fellow Dr. Rotin and @jacobpendergrast.bsky.social have made a substantial stride in dose modelling to target urine output onlinelibrary.wiley.com/doi/pdf/10.1... #medsky

011 1

Raza @transfusion.health · 24/04/2025

#NATA2025 The side for liberal hemoglobin transfusion thresholds make their case for MI and neurological injury #medsky

010 1

Raza @transfusion.health · 24/04/2025

Viscoelastic Testing in Post-Partum Hemorrhage Potentially useful, understudied HYPOcoagulability is important as well described HYPERcoagulability is also important, poorly studied #NATA25 #medsky #transfusion #obgyn

000 0

Raza @transfusion.health · 24/04/2025

Intravenous Iron at Home Dr. Elvira Bibs, reports on the Spanish experience Better for older patients, saves hospital beds, gives a high need treatment. Takeaway: select the right patients. Mind the logistics "What is lacking? Courage to implement it" #NATA25 #hemesky #medsky #transfusion

010 1

Raza @transfusion.health · 24/04/2025

Great talk to start off #NATA2025 by Dr. Paul Kalra, the investigator on IRONMAN trial for iron replacement in CHF → Patients feel better, and do better, with greater benefit in TSat >20% → No signal for increased infection related death No brainer! #medsky #hemesky #cardiology

000 0

Raza @transfusion.health · 19/04/2025

Neat abstract (and accompanying arXiv pre-print) on LLM agents for medical tasks at #neurips neurips.cc/virtual/2024... #ml #medsky

020 2

Raza @transfusion.health · 10/04/2025

We discuss the stats, causal inference, and visualization properties of transfusion probability curves in great a detail in the paper, and provide an R-tutorial to repeat our analyses on your data, available here: onlinelibrary.wiley.com/action/downl...

100 1

Raza @transfusion.health · 10/04/2025

The math can be repeated at each Hb result to give a transfusion probability curve to study and compare lab-guided practice The method can also be easily extended to other lab-guided clinical decisions, such as troponin → cardiac intervention, electrolyte level → replacement, and so on

100 1

Raza @transfusion.health · 10/04/2025

We introduce Transfusion Probability, calculated as the proportion of results of a certain value (or range) which were followed by a transfusion For example, if a physician saw 50 results of Hb = 7.0 g/dL, and transfused after seeing 15/50 results, transfusion probability is 30%

100 1

Raza @transfusion.health · 10/04/2025

Traditional methods to study the effect of labs on decisions use the last available lab result just before the decision, e.g. pre-trasfusion hemoglobin (PreHb) But PreHb only looks at lab values followed by a transfusion and ignores the denominator of the ones that weren't. Can be misleading!

100 1

Raza @transfusion.health · 10/04/2025

Lab values are strong drivers of medical decisions. Can we rigorously study how labs affect clinical behavior? We describe "Transfusion Probability"––a simple yet powerful way to study how labs influence the clinical decision to transfuse onlinelibrary.wiley.com/doi/10.1111/... #hemesky #stats

320 3

Raza @transfusion.health · 08/04/2025

Clinical guidelines can do no good sitting on shelves In @aabbupdates.bsky.social journal, co-authors and I argue that guidelines should come with an intentional plan for real-world uptake. We leverage #ImplementationScience to outline a process for doing so onlinelibrary.wiley.com/doi/10.1111/...

022 2

Raza @transfusion.health · 01/04/2025

Beautiful interactive visualizations of many Bayesian sampling algorithms. Makes the math come alive! #stats #bayesian chi-feng.github.io/mcmc-demo/ap...

010 1

Raza @transfusion.health · 31/03/2025

FARES-II: Much anticipated randomized trial @jama.com comparing Plasma to PCC in cardiac surgery (n=420) Those in PCC arm had ↑ hemostatic efficacy and ↓ RBC transfusion Hopefully practice-changing for many! jamanetwork.com/journals/jam... #transfusion #hemesky

000 0

Raza @transfusion.health · 24/03/2025

Publication bias eats Central Limit Theorem for breakfast... #stats #medsky onlinelibrary.wiley.com/doi/full/10....

091 9

Raza @transfusion.health · 21/03/2025

Ways of expressing quantitative information visually, from the amazing book "Visualization Analysis & Design" by Tamara Munzner #dataviz #stats

051 5

Raza @transfusion.health · 03/02/2025

Nice study at @aabbupdates.bsky.social journal, answers a sharp clinically relevant question onlinelibrary.wiley.com/doi/full/10....

011 1

Raza @transfusion.health · 01/02/2025

Good read here on this issue www.mdrc.org/sites/defaul...

010 1

Raza @transfusion.health · 25/01/2025

Interesting read in @naturecomms.bsky.social on meta-cognitive deficits in LLMs, with a good discussion of cognitive biases. Need more sophisticated, formal eval frameworks for biases in clinical decision-making www.nature.com/articles/s41...

000 0

Raza @transfusion.health · 24/01/2025

According to very smart allergy colleagues, we should probably stop using 1st Gen anti-histamines (like Benadryl) in transfusion and hematology (some reasoning below) Feels like a knowledge translation gap #medsky #hemesky aacijournal.biomedcentral.com/articles/10....

181 8

Raza @transfusion.health · 16/01/2025

Spline regression is incredibly relevant for medicine, not because biology has discontinuities, rather that we create them via thresholds on which we base medical decisions e.g. the definition of fever, neutropenia, age of insurance eligibility, transfusion thresholds, and so on.. #stats #medsky

2101 10

Raza @transfusion.health · 14/01/2025

How do cold-stored platelets(CSP) compare with room temp plts (RTP) for blood bank inventory? I created a fun sim of this in #RShiny. It's oversimplified, but good for understanding how hospitals with low plt usage could improve inventory with CSP > RT #transfusion razas3.shinyapps.io/CSP_Sim_R/

simulation of csp vs room temp platelets

120 2

Raza @transfusion.health · 09/01/2025

Sharing what was for me a highlight of last year's AABB conference The upcoming guideline recommendation for platelet threshold in lumbar puncture (>20 × 10e9/L) #medsky #transfusion

020 2

Raza @transfusion.health · 02/01/2025

Nice article in Nature Medicine on an evaluation framework for clinical LLM history-taking. A limitation is that they aren't using actual user clinical inputs. Their agentic framework is pretty neat for automating evaluations www.nature.com/articles/s41... #medsky #mlsky

030 3

Raza @transfusion.health · 02/01/2025

The Manifold Hypothesis is one of the most beautiful ideas I’ve encountered in a while, with deep intuitive appeal and explanatory power. Not sure if it’s a theory of everything in modeling but does make you seriously question all the naive models we tend to quibble about in medicine #mlsky

100 1

Raza @transfusion.health · 01/01/2025

"Overtransfusion" (OT) now has a consensus definition for hemovigilance and research OT is a key concept for understanding waste and harm associated with blood transfusion Table below summarizes ways in which OT arises onlinelibrary.wiley.com/doi/10.1111/...

120 2

Raza @transfusion.health · 28/12/2024

Should we eliminate leukocyte reduction for whole blood? It's an intriguing concept with mainly logistical benefits In @aabbupdates.bsky.social journal, my brilliant co-authors argue it's premature, and removal of LR deserves deeper, empirical scrutiny onlinelibrary.wiley.com/doi/10.1111/...

141 4

Raza @transfusion.health · 28/12/2024

The people in the middle

000 0

Raza @transfusion.health · 19/12/2024

The ICTMG 2024 Albumin Guidelines + resources + slide deck to help implement the guidelines at your local institutions + an explanatory webinar by yours truly All available here under "Use of Intravenous Albumin Guidelines": profedu.blood.ca/en/transfusi... #Transfusion #Hemesky

000 0

Raza @transfusion.health · 17/12/2024

Our correspondence with @natureportfolio.bsky.social advocating that scholarly journals support direct link-sharing to #Bluesky amidst the exodus of academic communities from X (formerly Twitter) @drgsbooth1.bsky.social @jwjacobs12.bsky.social @path-brian.bsky.social www.nature.com/articles/d41...

Screenshot of correspondence with @natureportfolio.bsky.social advocating that all scholarly journals support direct link-sharing to Bluesky amidst the exodus of many academic communities from X (formerly Twitter)

153 5

Raza @transfusion.health · 07/12/2024

Thanks Jarvis! You're right--no way to get around it for dashboards/recurring reports. Re: compound plots I just meant multiple plots within one panel--R can be finicky with plot areas/ratios I too have spent many hours struggling with network diagrams parameters until giving up altogether...

220 2

Raza @transfusion.health · 05/12/2024

Tranexamic acid did not prevent bleeding or increase clotting in this ~600 patient RCT of heme malignancy patients (TXA was given 1 g IV every 8 hours or 1.5 g orally every 8 hours)🤷🏽‍♂️ Which is too bad--we don't have a tonne to offer these patients.. :( www.thelancet.com/journals/lan... #transfusion

051 5

Raza @transfusion.health · 04/12/2024

Really cool application of convolutional neural networks for genomic analysis predicting blood group antigens from imputed genomic data @aabbupdates.bsky.social journal onlinelibrary.wiley.com/cms/asset/49...

040 4

Raza @transfusion.health · 04/12/2024

Great primer on Federated Learning (FL) methods and their potential for enabling large collaborative data analyses in transfusion medicine. Crucial for ML-worthy sample sizes Key idea: Bring the model to the data, not the data to the model onlinelibrary.wiley.com/doi/10.1111/... #MLSky

122 2

Raza @transfusion.health · 03/12/2024

Several image processing steps, 6 hrs of OCR, permuting embedding models and chunk sizes, creative prompt engineering, and many Flask deployments later (makes you miss RShiny)... ... and I can finally have a civil conversation with my favourite transfusion textbook of all time #mlsky #transfusion

110 1

Raza @transfusion.health · 01/12/2024

Video games have matured into one of the greatest art forms of this era. Pictured: Gris, which uses stunning artistic motifs to untangle despair in the aftermath of losing a loved one

000 0

Raza @transfusion.health · 01/12/2024

Obsidian Webclipper + AI = fantastic custom article summaries. Works for scientific literature too. Demo of how blazing fast it is @obsidian.md #ai

140 4

Raza @transfusion.health · 01/12/2024

Wonderful article by my mentor showing cognitive biases persist in AI chatbots, replicates known studies on biases in human and AI LLMs for comparison. Evidence that LLMs are inheriting some of our biases, in some cases quite strongly so ai.nejm.org/doi/full/10.... #mlsky #ai

041 4

Raza @transfusion.health · 30/11/2024

Accidental erythrocytes and spherocytes in the Christmas section #hemesky

090 9

Raza @transfusion.health · 29/11/2024

Do you wonder how a blood donation turns into a transfusable product? A new resource I helped develop for Canadian Blood Services gives an extensive walkthrough of how blood donations become blood products--for medical and curious non-medical folks profedu.blood.ca/en/VirtualTo... #transfusion

010 1