Digital twin predicts anatomy changes and contours for proton radiotherapy
A digital-twin framework for forecasting treatment-day imaging with contour uncertainty in adaptive proton radiotherapy
Artificial Intelligence
Summary
Treating head and neck cancer with proton therapy takes several weeks, during which the patient's anatomy changes but can't be directly imaged every day. The authors created a 'digital twin,' a computer model that predicts what the patient's body looks like on any treatment day based on past patient data and early scans. This twin also estimates how uncertain these predictions are, helping doctors understand how much the shape and position of tissues may vary. Their method can improve treatment targeting by anticipating anatomical changes without extra daily imaging.
What this means in practice
- •For radiation therapy teams: predict patient anatomy changes during proton treatment courses to adapt plans without daily imaging or manual contouring
- •For medical imaging software developers: integrate a patient-specific digital twin framework to estimate anatomy and contour uncertainty from limited early scans in therapy workflows
Authors
Yizhou Wu, Jie Ding, Justin Roper, Minglei Kang, Yuheng Li, Sibo Tian, David S. Yu, Xiaofeng Yang, Chih-Wei Chang
Abstract
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.