Session Item

This study evaluates the performance of an AI to automate beam angle selection for liver lesions treated with scanned proton beams. 

Forty-nine patients that received liver treatments between 2017 and 2020 with proton pencil beam scanning were divided into a training (n=31), validation (n=7), and test set (n=11) for the AI (Fig. 1a). The AI is based on casting beam-angle selection as a multi-label classification problem. To account for angular boundary discontinuity, the underlying convolution neural network was trained with a novel Circular Earth Mover’s Distance based regularization and multi-label circular-smooth label technique. After training, the network generates a prediction for the probability of each of the 72 classes that represent 5° steps in beam angles (Fig. 1b). The final AI angles are then produced by post-processing with an analytical algorithm that emulates proton planning clinical practice. I.e., the probability is modified for each beam angle based on the distance to the target and organs-at-risk occluding the target (Fig 1c). Performance was evaluated by comparing AI beam angles with the ground truth, i.e., angles selected by human planners.  Additionally, knowledge-based treatment planning was employed to automatically create treatment plans and assess the impact of beam angle choice on several dosimetric parameters.

For 8 of the 11 cases in the test set, AI-selected beam angles agreed with those determined by human planners to within 20 degrees (median angle difference = 10°; mean = 18.6°). Moreover, out of the total 22 beam angles predicted by the model, 15 (68%) were within 10 degrees of the human-selected angles (see table). 

The high correlation of human and AI predicted beam angles resulted in comparable plans in terms of dosimetric parameters (Fig 2). Select cases showed significant angle differences, dosimetric differences and their implications to overall plan quality will be discussed in terms of plan robustness, distal stoppage, and optimization settings.