Anticipatory Digital Twins for Online Head-and-Neck Adaptive Proton Therapy via Foundation-Model Registration

2026-08-01

AI Summary

In head-and-neck proton therapy, anatomical changes such as tumor shrinkage or weight loss can cause Bragg peak shifts, leading to underdosing of the target or overdosing of organs at risk. Current online adaptive radiotherapy relies on repeat CT scans and time-consuming replanning. This work proposes the first digital twin framework based on a pretrained foundation model that requires no patient-specific training. Leveraging a two-stage cross-patient deformable registration, it transfers longitudinal anatomical variations from population data to prospectively predict daily CT images and contours without same-day imaging. Evaluated on 88 patients, the predicted CTs improved normalized mutual information by 22.8%, increased organ-at-risk Dice coefficients by 20.2%, and reduced CT number errors by 23.4% compared to static planning CTs, with particularly pronounced benefits in patients exhibiting substantial anatomical changes.

Abstract

Head-and-neck (HN) proton therapy is highly sensitive to anatomical change over a 4-to-6-week course, as tumor shrinkage, weight loss, and setup variation can misposition the Bragg peak near critical organs such as the parotids, oral cavity, brainstem, and spinal cord, leading to target underdosing or organ-at-risk overdosing. Online adaptive proton therapy replans on the anatomy of the day, yet standard workflows rely on offline replanning that requires repeated CT acquisition and roughly a week of preparation, adding burden, cost, and delay. We investigate whether a patient's treatment-day anatomy can be predicted before image acquisition by transferring longitudinal change from a population database. We propose a digital-twin framework built on a pretrained foundation-model deformable registration network used without patient-specific training. A first registration aligns a prior patient's planning CT to the target and carries the prior's during-treatment quality assurance CT (QACT) into the target frame; a second registration estimates the prior's planning-to-QACT change, which is then applied to the target's own planning CT to synthesize predicted CTs (pdCTs) with propagated contours. Using 88 HN patients, each with a planning CT and three QACTs, we show that pdCTs better match treatment-day anatomy than the static planning CT. Compared with the planning CT alone, normalized cross-correlation improves by 22.8%, Dice for organs-at-risk by 20.2%, and CT-number error decreases by 23.4%. Gains are largest for patients with major anatomical change and negligible when anatomy is stable. This cross-patient motion transfer leverages the digital-twin concept to anticipate treatment-day anatomy, enabling personalized online adaptive proton therapy without repeated imaging.

Problem

• adaptive proton therapy
• anatomical change
• online replanning
• digital twin
• head-and-neck cancer

Innovation

• Digital Twin
• Foundation Model
• Deformable Registration
• Online Adaptive Proton Therapy
• Anatomical Change Prediction