Agenda

09:00 – 09:10
Harald Binder
Opening Remarks

Session 1: Similarity
Session Chair: Max Behrens

09:10 – 09:30
Nadine Binder & Jan Hasenauer
AI-assisted continuous-time modeling of metastatic breast cancer under sparse and irregular observation

09:30 – 09:50 
Maria Kalweit & Evelyn Ullrich
BLINK: Modeling NK cell cytotoxicity as a latent dynamical process from time-resolved fluorescence microscopy

09:50 – 10:30 
Panel Discussion: The role of similarity in small data
Panelists: Nadine Binder, Jan Hasenauer, Maria Kalweit, Evelyn Ullrich

10:30 – 11:00 
Break

Session 2: Transfer
Session Chair: Julia Hindel

11:00 – 11:25 
Frank Hutter & Pascal Schlosser
Advancing Tabular Foundation Models for longitudinal data integration

11:25 – 11:50 
Anna Köttgen & Johannes Hertel
From population studies to whole body modeling and back

11:50 – 12:30 
Early career researcher flash presentations on collaborative work

12:30 – 13:30 
Lunch & poster session

Session 3: Uncertainty
Session Chair: Carola Heinzel

13:30 – 13:50 
Heinz Wiendl & Susanne Weber
Clinical categories to statistical trajectories: An interdisciplinary approach to modeling disease progression in multiple sclerosis

13:50 – 14:10 
Martin Wolkewitz & Maria Elena Maccari
Challenges of target trial emulation in rare pediatric diseases: from limited data to clinical management

14:10 – 14:50 
Panel Discussion: The role of uncertainty in small data
Panelists: Heinz Wiendl, Martin Wolkewitz, Maria Elena Maccari, Peter Pfaffelhuber

14:50 – 15:00 
Harald Biner
Outlook on the small data landscape

Agenda

This time, STRATOS is organizing a mini-symposium for two different target groups on October 1st.

The morning session (09:00–12:30, Methodological aspects) is designed in the usual format of a half-day meeting with several shorter talks on behalf of TGs and panels.

In the afternoon (13:30–17:00, Practical aspects of medical research),we offer, for the first time, a workshop for clinicians and data analysts with less training and experience in statistical topics. Gary Collins (University of Birmingham, UK) and Matthias Briel (University Hospital Basel, Switzerland) will give keynote presentations.

 

Session 1: Methodological Aspects

9:00 – 10:30 and 11:00-12:30

Session 1 A
Session Chair: Willi Sauerbrei

9:00 – 9:10
Willi Sauerbrei
Introduction

9:10 – 9:35
James R. Carpenter
P-values and hypothesis testing: beyond polemics to practical solutions

9:35 – 10:00
Michal Abrahamowicz
How important are the Hazards of Hazard Ratios?

10:00 – 10:25
Anne C. M. Thiébaut
Methods for Adjusting for Covariate Measurement Error in Flexible Modelling of Functional Form: Results of a Blinded, Controlled Neutral Comparison Simulation Study

Session 1B
Session Chair: Georg Heinze

11:00 – 11:25
Marianne Huebner
Statistical analysis plan with initial data analysis (SAPI): Validating the SAPI checklist in analysis projects

11:25 – 11:50
Doranne Thomassen
Comparing five frameworks that may be used to define estimands: friends not foes

11:50 – 12:15
Els Goetghebeur
Sensitivity analyses for missing data in observational studies: A practical guide for planning, conducting and reporting under MAR and beyond

12:15 – 12:30
Discussion 1 (Chair: Georg Heinze)

 

Session 2: Practical Aspects of Medical Research

13:30 – 15:00 and 15:30-17:00

Session 2A
Session Chair: Willi Sauerbrei 

13:30 – 13:40
Willi Sauerbrei
Introduction

13:40 – 14:10
Gary Collins
The importance of methodology and transparency in predictive AI

14:10 – 14:27
Riccardo De Bin
From traditional statistical modelling to machine learning algorithms: a perspective from the STRATOS TG9 high-dimensional data viewpoint

14:27 – 14:44
Ben van Calster
Performance evaluation of predictive AI models to support medical decisions: overview and guidance

14:44 – 15:01
Peggy Sekula
Designing Prognostic Factor Studies: Key Concepts and Practical Guidance

Session 2B
Session Chair: Anne-Laure Boulesteix

15:30 – 16:00
Matthias Briel
Meta-research to improve evidence generation and synthesis

16:00 – 16:15
Theresa Ullmann
The Problem with Univariable Selection in Regression Modelling — and What to Do Instead

16:15 – 16:30
Saskia le Cessie
Guidelines to the design, analysis and interpretation of patient-reported outcomes in cancer clinical trials

16:30 – 16:45
Willi Sauerbrei
An overview and categorization of papers published in statistical series of medical journals

16:45 – 17:00
Discussion 2 (Chair: Anne-Laure Boulesteix)

Agenda

Session 1

09:00 – 09:15
Opening remarks

09:15 – 10:00
Kristy Robledo
What Does AI Mean for the Future of Biostatistics?

10:00 – 10:30
Speed Networking

10:30 – 11:00

Coffee Break

Session 2

11:00 – 11:30
Jasko Miebs
Hallucination Detection in Large Language Models

11:30 – 12:00
Speaker 3 TBD

12:00 – 12:30
Dilemma Game

12:30 – 13:00
Lunch Break

Session 3

13:30 – 14:30
Interactive Session

14:30 – 15:00
Panel Discussion – Voices from Early Career Researchers

Open speakers to be announced in due course.

Session Descriptions

Speed Networking

Attendees are divided into groups of about ten. Each participant has two to three minutes to give an “elevator pitch” or “three-minute thesis” style introduction of themselves and their work or research. A great opportunity to showcase your work while making new connections.

Dilemma Game

Building on the success of last year’s session, this year’s ECR Day includes an interactive session where all participants discuss the challenges statisticians face in conducting biostatistical research. We use the Dilemma Game app, developed by Erasmus University Rotterdam (EUR), to pose questions and scenarios related to professionalism and integrity in research for discussion. We recommend that all attendees download the app before the mini-symposium.

• Learn more: https://www.eur.nl/en/about-university/policy-and-regulations/integrity/research-integrity/dilemma-game
• iOS: https://apps.apple.com/us/app/dilemma-game/id1494087665
• Android: https://play.google.com/store/apps/details?id=nl.eur.dilemmagame&hl=nl

Panel Discussion – Voices from Early Career Researchers

A key aim of the ECR Day is to provide a forum for discussing the challenges biostatisticians face. To support this, we are introducing a new panel discussion that places early-career researchers at the center of the conversation.

Rather than the traditional format where established researchers share their experiences with a largely listening audience, this session gives early-career researchers an opportunity to discuss the opportunities, challenges, and realities of building a career in statistics and biostatistics today. Topics may include research and collaboration, publishing, career development, balancing competing demands, and navigating transitions between academia, industry, and healthcare.

The session aims to encourage open discussion, shared experiences, and peer learning, reflecting the collaborative spirit of ECR Day.

We invite all delegates registered for ECR Day who identify as early career researchers to put themselves forward to join the panel. A range of perspectives and backgrounds is welcome, and no previous panel experience is required. Please register your interest here: https://forms.office.com/