Purpose: To develop a computational algorithm based on an artificial neural network (ANN) that allows treatment verification using an EPID in cine mode for hypofractionated lung radiotherapy. Method and Materials: We developed a novel ANN based technique using cine EPID images to verify that the target was within the beam aperture when the beam was on. The first step of using the ANN involved training from a training dataset. We simulated training images, i.e., cine EPID images with different tumor locations, by shifting DRRs relative to the beam aperture. With a pre‐defined threshold p%, we associated category 1 to the training image if more than p% of the tumor projection in the beam eye view was within the aperture and category −1 otherwise. The trained network could therefore analyze the cine EPID images obtained during the treatment and classify them into the corresponding category 1 or −1. Results: Two patients, each treated with 5 fractions, were included in our feasibility study. A radiation oncologist read the cine EPID images and classified them into category 1 or −1; this served as our ground truth. The ANN was applied to the training images to build the neural network. We set p%=95% for this study. For each treatment field, one neural network needs to be built. Averaging over both patients and all fields, the trained network successfully classified 97.5% of the cine EPID images overall. Conclusion: The proposed ANN based technique can successfully analyze cine EPID images to verify whether or not the tumor is within the beam aperture, and it can do so with high accuracy. This technique provides an important clinical safeguard—whenever the tumor moves out of the irradiation field, the treatment beam can be interrupted, so that radiation won't be unnecessarily delivered to normal tissues.
ASJC Scopus subject areas
- Radiology Nuclear Medicine and imaging