Dynamic Modulated Arc Therapy (DMAT): An Intent-Driven, Time-Aware Framework for Next-Generation Radiotherapy Delivery

TL;DR

DMAT is a novel, intent-driven framework for radiotherapy that co-optimizes dosimetric quality, delivery time, and modulation complexity by integrating machine dynamics and adaptive control.

Contribution

Introduces DMAT, a framework that jointly optimizes treatment quality, time, and complexity considering machine dynamics and clinical cost functions.

Findings

Higher modulation increases MU/Gy and delivery time.

Adaptive CP allocation improves dose quality in high-reward sectors.

Prioritizing efficiency reduces modulation and delivery time with minor quality loss.

Abstract

Traditional VMAT optimization often ignores dynamic machine limits, treating delivery time as an emergent property rather than a steerable parameter. This work introduces Dynamic Modulated Arc Therapy (DMAT), an intent-driven framework that jointly co-optimizes dosimetric quality, delivery time, and modulation complexity. DMAT couples machine emulation accounting for axis synchronization and finite acceleration with dynamic modulation control and clinical cost functions. A user-selected level (-3 to +3) governs leaf-travel, MU behavior, and CP density. Plans are created by initializing CP geometry, leaf positions, and MU, then iteratively alternating dosimetric updates with sequencing updates (penalties for motion, velocity changes, MU uniformity, and complexity), followed by post-processing. CP density is adapted by first optimizing with a uniform distribution and then redistributing CPs to arc sectors with higher complexity. DMAT was evaluated using a hypothetical system (2.5 RPM gantry, 6.25 cm/s MLC, 3000 MU/min) on H&N, lung SBRT, and prostate SBRT cases. Higher modulation levels produced increased MU/Gy and longer delivery times, while adaptive CP allocation concentrated resolution in high-reward sectors. H&N cases showed substantial quality gains with increased modulation, whereas prostate and lung SBRT exhibited smaller incremental improvements. When efficiency was prioritized (negative levels), DMAT reduced modulation and maintained a constant CP budget while shortening delivery time, producing quantifiable reductions in plan quality. DMAT enables intent-driven planning where quality and complexity are co-optimized via machine-aware timing. Accurate delivery time is exposed during planning, making trade-offs transparent and navigable for next-generation systems and time-constrained workflows like motion-sensitive or adaptive radiotherapy.