Abstract
Randomized controlled trials (RCTs) are a critical part of evidence-based medicine — they generate rigorous evidence supporting the efficacy of medical treatments at the population level. There has been much interest in performing RCTs in silico by creating a digital twin of each individual in a study population and using the twin to predict the effect of one or more candidate treatments of interest. However, this apparently straightforward approach relies on high-fidelity digital twins that are not usually available; indeed, thorough validation of digital human twins in interventional settings already requires RCT data. Here, we discuss a different approach: in cohort-level twinning, our goal is to create a digital representation of an entire patient cohort’s baseline and response profiles, but not necessarily of each patient in the cohort itself. Cohort-level twinning allows us to leverage coarse-grained models such as ordinary differential equations and agent-based simulations to perform in silico trials that can be used to plan RCTs and make predictions about the population-level effects of novel interventions. We illustrate cohort-level twinning using an agent-based simulation of a metastasizing tumor implemented using a cellular Potts model. We fit this simulation to reproduce progression-free survival times in a cohort of non-small-cell lung cancer patients, and demonstrate how such a fitted model can be used to conduct an in silico trial and a power analysis.
Links to Paper and Supplementary Materials
Main Paper (Open Access Version): https://papers.miccai.org/miccai-2026-sat/paper/DT4H_023.pdf
SharedIt Link: Not yet available
SpringerLink (DOI): Not yet available
Supplementary Material: Not Submitted
Link to Open Review
Open Review Page: https://openreview.net/forum?id=W9NmIfwBPx
BibTex
@InProceedings{SchJan_CohortLevel_MICCAISAT2026,
author = { Schering, Jan AND Wouters, Lin AND Wortel, Inge AND Textor, Johannes},
title = { { Cohort-Level Twinning in an Agent-Based Model of Metastatic Tumor Growth } },
booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026 Workshops and Challenges},
year = {2026},
publisher = {Springer Nature Switzerland},
volume = {LNCS 17275},
month = {pending},
page = {pending}
}