“Deep Learning Improves Detection and Reduces False Positives in In Vivo Dosimetry Using Electronic Portal Imaging Devices”

https://pubmed.ncbi.nlm.nih.gov/38112213

This study found that deep learning-based tools can improve the detection of plan-specific deviations and reduce false positives in electronic portal imaging device-based in vivo dosimetry by reconstructing underlying patterns of generic deviations in gamma-distributions.