Fully-automated deep learning-powered system for DCE-MRI analysis of brain tumors
作者：
Highlights：
• We introduce the first fully-automated system for DCE-MRI analysis of brain tumors.
• We introduce the algorithms for tumor and vascular input region segmentation.
• We introduce a cubic model of the vascular input function for pharmacokinetic modeling.
• We thoroughly validate our method using benchmark and clinical data.
• We show that our method obtains state-of-the-art and reproducible results.
• Our system requires less than 3 min for the end-to-end processing using one GPU.
摘要
•We introduce the first fully-automated system for DCE-MRI analysis of brain tumors.•We introduce the algorithms for tumor and vascular input region segmentation.•We introduce a cubic model of the vascular input function for pharmacokinetic modeling.•We thoroughly validate our method using benchmark and clinical data.•We show that our method obtains state-of-the-art and reproducible results.•Our system requires less than 3 min for the end-to-end processing using one GPU.
论文关键词：Deep neural network,Pharmacokinetic model,Tumor segmentation,DCE-MRI,Perfusion,Brain
论文评审过程：Received 5 November 2018, Revised 28 October 2019, Accepted 20 November 2019, Available online 27 November 2019, Version of Record 4 December 2019.
论文官网地址：https://doi.org/10.1016/j.artmed.2019.101769
