AI digital twin plans liver tumor ablation 420x faster, cuts organ damage 55%

Microwave ablation (MWA) is a minimally invasive treatment for liver tumors, but its success hinges on meticulously planning the antenna insertion trajectory, power, and duration. Traditional numerical simulation offers reliable predictions but is computationally expensive, making optimization-based planning impractical. To solve this, a team of researchers created a digital twin-based automatic planning framework that pairs a neural ablation prediction model with a genetic algorithm. The model was trained on multiphysics simulation data generated from patient-specific tumor and vessel structures, antenna configurations, and treatment conditions, enabling it to act as a fast forward model during the optimization process.

The results are striking: the prediction model achieved a Dice score of 95.1%, meaning its predicted ablation zones closely match ground truth. In 13 previously unseen planning cases, the framework improved ablation efficiency by 54.3% and reduced organ damage by 55.0% compared with clinician-defined plans, even slightly shortening the insertion path length by 3.3%. Most generated plans were judged clinically applicable by MWA specialists. Crucially, the entire planning process runs approximately 420-fold faster than numerical-simulation-based planning, turning what once took minutes to hours into near-instantaneous, data-driven decisions. The code is openly available on GitHub, allowing other researchers and clinicians to replicate and build upon this work.