Enhancing Radiotherapy for Locally Advanced Non-Small Cell Lung Cancer Patients with iCE, a Novel System for Automated Multi-Criterial Treatment Planning Including Beam Angle Optimization

In this study, the novel iCE radiotherapy treatment planning system (TPS) for automated multi-criterial planning with integrated beam angle optimization (BAO) was developed, and applied to optimize organ at risk (OAR) sparing and systematically investigate the impact of beam angles on radiotherapy dose in locally advanced non-small cell lung cancer (LA-NSCLC). iCE consists of an in-house, sophisticated multi-criterial optimizer with integrated BAO, coupled to a broadly used commercial TPS. The in-house optimizer performs fluence map optimization to automatically generate an intensity-modulated radiotherapy (IMRT) plan with optimal beam angles for each patient. The obtained angles and dose-volume histograms are then used to automatically generate the final deliverable plan with the commercial TPS. For the majority of 26 LA-NSCLC patients, iCE achieved improved heart and esophagus sparing compared to the manually created clinical plans, with significant reductions in the median heart Dmean (8.1 vs. 9.0 Gy, p = 0.02) and esophagus Dmean (18.5 vs. 20.3 Gy, p = 0.02), and reductions of up to 6.7 Gy and 5.8 Gy for individual patients. iCE was superior to automated planning using manually selected beam angles. Differences in the OAR doses of iCE plans with 6 beams compared to 4 and 8 beams were statistically significant overall, but highly patient-specific. In conclusion, automated planning with integrated BAO can further enhance and individualize radiotherapy for LA-NSCLC.

Exploring Lifted Planning Encodings in Essence Prime

State-space planning is the de-facto search method of the automated planning community. Planning problems are typically expressed in the Planning Domain Definition Language (PDDL), where action and variable templates describe the sets of actions and variables that occur in the problem. Typically, a planner begins by generating the full set of instantiations of these templates, which in turn are used to derive useful heuristics that guide the search. Thanks to this success, there has been limited research in other directions. We explore a different approach, keeping the compact representation by directly reformulating the problem in PDDL into ESSENCE PRIME, a Constraint Programming language with support for distinct solving technologies including SAT and SMT. In particular, we explore two different encodings from PDDL to ESSENCE PRIME, how they represent action parameters, and their performance. The encodings are able to maintain the compactness of the PDDL representation, and while they differ slightly, they perform quite differently on various instances from the International Planning Competition.