Toward Actionable Digital Twins for Radiation-Based Imaging and Therapy: Mathematical Formulation, Modular Workflow, and an OpenKBP-Based Dose-Surrogate Prototype

Hsin-Hsiung Huang; Bulent Soykan

eess.IVMar 26, 2026

Toward Actionable Digital Twins for Radiation-Based Imaging and Therapy: Mathematical Formulation, Modular Workflow, and an OpenKBP-Based Dose-Surrogate Prototype

Authors: Hsin-Hsiung Huang, Bulent Soykan

Organizations: School of Data, Mathematical, and Statistical Sciences, University of Central Florida, Orlando, FL 32816 USA · Institute for Simulation and Training, University of Central Florida, Orlando, FL 32826 USA

Abstract

Digital twins for radiation-based imaging and therapy are most useful when they assimilate patient data, quantify predictive uncertainty, and support clinically constrained decisions. This paper presents a modular framework for actionable digital twins in radiation-based imaging and therapy and instantiates its reproducible open-data component using the \openkbpfull{} benchmark. The framework couples PatientData, Model, Solver, Calibration, and Decision modules and formalizes latent-state updating, uncertainty propagation, and chance-constrained action selection. As an initial implementation, we build a GPU-ready PyTorch/MONAI reimplementation of the \openkbp{} starter pipeline: an 11-channel, 19.2M-parameter 3D U-Net trained with a masked loss over the feasible region and equipped with Monte Carlo dropout for voxel-wise epistemic uncertainty. To emulate the update loop on a static benchmark, we introduce decoder-only proxy recalibration and illustrate uncertainty-aware virtual-therapy evaluation using DVH-based and biological utilities. A complete three-fraction loop including recalibration, Monte Carlo inference, and spatial optimization executes in 10.3~~s. On the 100-patient test set, the model achieved mean dose and DVH scores of 2.65 and 1.82~~Gy, respectively, with 0.58~s mean inference time per patient. The \openkbp{} case study thus serves as a reproducible test bed for dose prediction, uncertainty propagation, and proxy closed-loop adaptation, while future institutional studies will address longitudinal calibration with delivered-dose logs and repeat imaging.

Figures & tables

Module	Radiation-based imaging and therapy instantiation
PatientData	Harmonized imaging, structures, plan goals, dose distributions, delivered logs, metadata, and outcomes.
Model	Hybrid physics and learned models, including dose surrogate, accumulated dose, response, and toxicity models.
Solver	Forward prediction for dose and downstream summaries, with sensitivities when needed for calibration and decision.
Calibration	Sequential updating using new measurements, implemented as inverse problems or filtering as in ( 4 )-( 5 ).
Decision	Constrained protocol selection using predictive distributions, utilities, and safety constraints as in ( 13 )-( 14 ).

TABLE I: Module mapping from a modular digital twin workflow to radiation-based imaging and therapy.

Panel of Figure 2 — Fig. 1: Actionable digital twin architecture for radiation-based imaging and therapy. Measurements update PatientData, calibration updates model state or parameters, the solver produces predictive distributions with uncertainty, and the decision module selects constrained actions for the clinical system.

Aspect	TumorTwin	Proposed RT-DT prototype
Primary domain	General oncology digital twins with mechanistic tumor-growth and response models	External-beam radiation therapy with CT/ROI-driven volumetric dose-surrogate modeling
Primary observations	Longitudinal imaging and tumor-response measurements	Planning CT, structure masks, feasible dose mask, reference plan dose, and proxy fraction-level dose summaries
Calibration target	Patient-specific mechanistic parameters and latent tumor state	Dose-surrogate state or parameters via proxy sequential calibration, with encoder freezing and decoder adaptation
Decision output	Alternative-treatment forecasting and optimization	Uncertainty-aware plan selection or spatial modulation under RT-specific constraints
Time horizon	Medium- to long-term response forecasting	Short-term fractionated adaptation and dose-evaluation loop
Computational profile	Numerical solvers, gradients, and model-based calibration	Rapid deep-learning inference with Monte Carlo dropout and lightweight transfer-learning updates

TABLE II: Framework-level comparison between TumorTwin and the proposed radiation-therapy digital-twin (RT-DT) prototype.

Metric	Mean	Std. dev.
Dose score [Gy]	2.65	0.42
DVH score [Gy]	1.82	0.55
Inference time / patient [s]	0.58	0.05

TABLE III: Cohort-wide OpenKBP test-set benchmark metrics for the present model (100 patients).

Fig. 2: Representative axial slice for OpenKBP case pt_289: reference dose, predicted mean dose $\mu(\mathbf{x})$ , absolute error $|\mu(\mathbf{x})-d(\mathbf{x})|$ , and voxel-wise epistemic uncertainty $\sigma(\mathbf{x})$ estimated by Monte Carlo dropout.

Fig. 3: Uncertainty-aware DVHs for case pt_289. Solid curves denote reference DVHs, dashed curves denote predictive-mean DVHs, and shaded regions indicate pointwise 95% predictive intervals from the Monte Carlo dropout ensemble.

Fig. 4: Illustrative proxy virtual-therapy adaptation trajectory. Top: surrogate TCP and NTCP objectives across 30 fractions with a simulated anatomical shift at fraction 10. Bottom: corresponding evolution of the epistemic uncertainty summary

\sigma(\mathbf{x})

. The example demonstrates the response of the calibration and decision modules in a proxy experiment; it is not intended as a validated estimate of clinical outcome probabilities.

Radiotherapy Dose Prediction

See all 12 papers

A digital-twin framework for forecasting treatment-day imaging with contour uncertainty in adaptive proton radiotherapy
Sep 21, 2026Yizhou Wu, Jie Ding, Justin Roper +6Digital Twins Healthcare
Department of Radiation Oncology and Winship Cancer Institute, Emory University, Atlanta, GA 30308 · Department of Human Oncology, University of Wisconsin, Madison, WI 53792 · Department of Radiation and Cellular Oncology, University of Chicago, Chicago, IL 60637
A practical probabilistic framework for deformable image registration uncertainty in radiotherapy dose propagation
Jun 8, 2026Stefan Heldmann, Sven Kuckertz, Nasim Givehchi +4Uncertainty Quantification Oncology
Fraunhofer Institute for Digital Medicine MEVIS, Lübeck, Germany · Varian, a Siemens Healthineers company, Dättwil, Switzerland · Icon Group, South Brisbane, QLD, Australia
An Uncertainty-Guided Digital Twin Framework for Online Adaptive Proton Therapy in Head and Neck Cancer: A Feasibility Study
Sep 30, 2026Yizhou Wu, Ryan J. Sanford, Huiqiao Xie +12Healthcare Digital Twins
Department of Radiation Oncology and Winship Cancer Institute, Emory University, Atlanta, GA 30308, USA · Department of Radiation Oncology, Taipei Medical University Hospital, Taipei, 110301, Taiwan · Department of Human Oncology, University of Wisconsin, Madison, WI 53792, USA +1

Sep 21, 2026J/K move · Enter open · S save

A digital-twin framework for forecasting treatment-day imaging with contour uncertainty in adaptive proton radiotherapy

Yizhou Wu, Jie Ding, Justin Roper, Minglei Kang+5

Department of Radiation Oncology and Winship Cancer Institute, Emory University, Atlanta, GA 30308 · Department of Human Oncology, University of Wisconsin, Madison, WI 53792 · Department of Radiation and Cellular Oncology, University of Chicago, Chicago, IL 60637

Digital Twins Healthcare Segmentation Uncertainty Estimation

Open paper AskPDF arXiv

Head-and-neck anatomy changes over a six-to-seven-week proton course, and the anatomy of a later week cannot be imaged when the plan is made. We present a digital-twin framework that forecasts a patient's treatment-day anatomy as an ensemble of predicted CTs with propagated contours and quantifies the uncertainty of the forecast contours. The twin is a library of previously treated patients with planning and weekly quality-assurance CTs (QACTs), made patient-specific by a two-step foundation-model deformable registration: a cross-patient field carries each library patient onto the current patient, and a longitudinal field, estimated in the current patient's frame, carries that patient's planning-to-QACT change onto the current patient's own planning CT. A library of 302 observations from 88 patients yields about 300 replicates per patient, each a deformation that occurred in a treated patient. The dispersion of the propagated contours, resolved by outward normal, is six-direction contour uncertainty in millimeters. This is uncertainty in the input to the forecast, which library patient the current patient follows, rather than in model parameters, and it is unchanged when the registration engine is exchanged. On ten patients with clinician contours on two QACTs, the library alone fixes the anisotropic shape of the uncertainty (4.5 to 6.2 mm); the first QACT narrows it by a factor of 3.2 to 3.6 without a contour being drawn; an approved contour improves the center but not the width. The estimate orders directions correctly but is not Gaussian-calibrated. A clinical target volume expansion is worked out as one application.

Abstract

Radiotherapy Dose Prediction

Explore similar work

A digital-twin framework for forecasting treatment-day imaging with contour uncertainty in adaptive proton radiotherapy

A practical probabilistic framework for deformable image registration uncertainty in radiotherapy dose propagation

An Uncertainty-Guided Digital Twin Framework for Online Adaptive Proton Therapy in Head and Neck Cancer: A Feasibility Study