作者: Furen Xiao , Jen-Tang Lu , Feng-Ming Hsu , Shao-Lun Lu , Wei-Hung Weng
DOI:
关键词:
摘要: Stereotactic radiosurgery (SRS), which delivers high doses of irradiation in a single or few shots to small targets, has been standard care for brain metastases. While very effective, SRS currently requires manually intensive delineation tumors. In this work, we present deep learning approach automated detection and segmentation metastases using multimodal imaging ensemble neural networks. order address multiple metastases, further propose volume-aware Dice loss optimizes model performance the information lesion size. This work surpasses current benchmark levels demonstrates reliable AI-assisted system treatment planning