NewsFeed Voices Science Insight Innovation
January, 2026
January 2026
M T W T F S S
 1234
567891011
12131415161718
19202122232425
262728293031  
Doranne Thomassen on Quantifying Uncertainty in Machine Learning Models
Doranne Thomassen/LinkedIn
Dec 1, 2025, 10:23

Doranne Thomassen on Quantifying Uncertainty in Machine Learning Models

Doranne Thomassen, Postdoc researcher (Medical Statistics) at Leiden University Medical Center, shared on LinkedIn:

”New publication!

The Lancet Digital Health has published our Viewpoint on uncertainty in ML-based risk predictions in healthcare.

In this study, we developed a computational method to express individual prediction uncertainty as an effective sample size, with illustrations for logistic regression, elastic net, neural net, random forest and XGBoost.

Our findings emphasize the importance of considering prediction uncertainty at the individual level from the perspective of trustworthy AI.

Effective sample sizes can be used when evaluating the trustworthiness of predicted risks for individual patients.

Thanks to the team Saskia le Cessie, Ewout Steyerberg, Toby Hackmann Jelle Goeman, LUMC – Biomedical Data Sciences”

Read the full article in The Lancet Digital Health.

Article: Effective sample size for individual risk predictions: quantifying uncertainty in machine learning models

Authors: Doranne Thomassen, Toby Hackmann, Jelle Goeman, Ewout Steyerberg, Saskia le Cessie

Doranne Thomassen on Quantifying Uncertainty in Machine Learning Models

Stay updated on all scientific advances with Hemostasis Today.

AI Doranne Thomassen Ewout Steyerberg Health Hematology Hemostasis Hemostasis Today Jelle Goeman LUMC - Biomedical Data Sciences Medicine ML risk predictions Saskia le Cessie The Lancet Digital Health thrombosis Toby Hackmann
NewsFeed