AUTHOR=Fan Shengyu , Bian Yueyan , Chen Hao , Kang Yan , Yang Qi , Tan Tao TITLE=Unsupervised Cerebrovascular Segmentation of TOF-MRA Images Based on Deep Neural Network and Hidden Markov Random Field Model JOURNAL=Frontiers in Neuroinformatics VOLUME=Volume 13 - 2019 YEAR=2020 URL=https://www.frontiersin.org/journals/neuroinformatics/articles/10.3389/fninf.2019.00077 DOI=10.3389/fninf.2019.00077 ISSN=1662-5196 ABSTRACT=Automated cerebrovascular segmentation of time-of-flight magnetic resonance angiography (TOF-MRA) images can provide the quantitatively measuring method to evaluate the disorder of human cerebrovascular system. Deep neural network (DNN)-based cerebrovascular segmentation methods have shown to give an outstanding performance, but they are limited by huge training dataset. In this paper, we proposed an unsupervised cerebrovascular segmentation method of TOF-MRA images based on DNN and hidden Markov random field (HMRF) model. DNN-based cerebrovascular segmentation model is trained by the labeling of HMRF rather than manual annotations. The proposed method was trained and tested by 100 cases of TOF-MRA images. The result was evaluated by dice similarity coefficient (DSC), which reached to 0.79. The trained model had a better performance than the traditional HMRF-based cerebrovascular segmentation method in binary pixel-classification. This paper merges advantages of both DNN and HMRF to address the limitation of amounts of annotations during deep learning training, which is a more effective cerebrovascular segmentation method.